Subset (24)

qwen_core · 1.3M rows

Split (1)

collection · 1.3M rows

record_id stringlengths 69 86	task_id stringlengths 69 69	prompt_hash stringlengths 64 64	source stringclasses 13 values	source_repo stringclasses 13 values	source_revision stringclasses 13 values	source_file stringclasses 33 values	source_row_id stringlengths 1 49	source_split stringclasses 6 values	source_url stringclasses 33 values	license stringclasses 8 values	attribution stringlengths 48 344	pool stringclasses 2 values	prompt stringlengths 1 271k	context stringlengths 0 98.7k	family stringclasses 9 values	leaf stringclasses 8 values	taxonomy_status stringclasses 5 values	taxonomy_reason stringclasses 86 values	language stringclasses 16 values	rubric_origin stringclasses 2 values	reference_solution stringlengths 0 45.3k	source_metadata_json stringlengths 42 1.82M	source_record_sha256 stringlengths 64 64	upstream_task_id stringlengths 0 81	source_difficulty stringclasses 30 values	estimated_difficulty stringclasses 5 values	difficulty_basis stringclasses 28 values	difficulty_confidence stringclasses 4 values	core_mixture_recommendation stringclasses 4 values	empirical_difficulty stringlengths 12 896	validation_status stringclasses 1 value	source_representation stringclasses 9 values	record_kind stringclasses 2 values	source_row int64 0 299k	quality_flags listlengths 3 16	has_native_rubric bool 2 classes	has_reference_solution bool 2 classes	has_verifier bool 2 classes	prompt_context_id stringlengths 69 69	annotation_count int64 1 112	eligible_annotation_count int64 1 23	group_sources listlengths 1 4	source_split_status stringclasses 3 values	curation_status stringclasses 1 value	curation_reasons listlengths 0 0	difficulty_tier stringclasses 2 values	difficulty_evidence listlengths 2 5	difficulty_tier_confidence stringclasses 2 values	difficulty_target_model stringclasses 1 value	difficulty_validation stringclasses 1 value
gretel_sql:4414be351e3cb0325e4c01cfcd30bf7c54fa8936380626256464d49cf0967d7b	task:4bfe093341fa2e9148b3b3746378a75360e1ad1041add250e13ba8b29f731ce6	83018c64a584c8a72f9ef73d108c601f690a0a89d09cefb571b96fd3ea3239a6	gretel_sql	gretelai/synthetic_text_to_sql	740ab236e64503fba51be1101df7a1be83bf455d	synthetic_text_to_sql_test.snappy.parquet	25	test	https://huggingface.co/datasets/gretelai/synthetic_text_to_sql/resolve/740ab236e64503fba51be1101df7a1be83bf455d/synthetic_text_to_sql_test.snappy.parquet	apache-2.0	Gretel Synthetic Text-to-SQL dataset creators; source card and author citation preserved in the download manifest.	proposed	What is the total amount donated to each program?	CREATE TABLE Programs (ProgramID INT, ProgramName VARCHAR(255)); INSERT INTO Programs (ProgramID, ProgramName) VALUES (1, 'Education'), (2, 'Health'), (3, 'Environment'); CREATE TABLE DonorsPrograms (DonorID INT, ProgramID INT); INSERT INTO DonorsPrograms (DonorID, ProgramID) VALUES (1, 1), (2, 1), (3, 2), (4, 2), (5, ...	database_business_operations	database_business_operations.database_query_analysis	heuristic_unreviewed	Mapped from source sql_task_type='analytics and reporting'; no individual SQL execution or semantic review.	en	none	SELECT Programs.ProgramName, SUM(Donations.DonationAmount) AS TotalDonatedAmount FROM Programs INNER JOIN DonorsPrograms ON Programs.ProgramID = DonorsPrograms.ProgramID INNER JOIN Donations ON DonorsPrograms.DonorID = Donations.DonorID GROUP BY Programs.ProgramName;	{"domain":"nonprofit operations","domain_description":"Donation records, program outcomes, volunteer engagement metrics, budget reports, and community impact assessments.","id":25,"sql_complexity":"multiple_joins","sql_complexity_description":"two or more joins (specify inner, outer, cross)","sql_explanation":"The SQL ...	b8d2a0ae5ef1c7dd5edd627ad573a4aafdf3a31b91b9cd70013e13bd8393f078	gretel_sql:test:25	multiple_joins	intermediate	Estimated from source sql_complexity='multiple_joins'; query feature labels are only proxies for difficulty, not measured solver performance. No empirical solver calibration was performed.	heuristic	needs_difficulty_calibration	not_measured	analysis_only_unvalidated	primary	task_annotation	24	[ "difficulty_not_calibrated", "not_individually_reviewed", "reference_not_independently_verified", "reference_or_constraint_only_no_rubric", "response_diversity_not_validated", "source_evaluation_split", "sql_dialect_and_result_equivalence_need_review", "sql_not_executed", "synthetic_reference" ]	false	true	false	task:4bfe093341fa2e9148b3b3746378a75360e1ad1041add250e13ba8b29f731ce6	1	1	[ "gretel_sql" ]	held_out	curated_candidate	[]	core_candidate	[ "source_sql_complexity:multiple_joins", "sql_structure:selects=1;joins=2;statements=1;advanced=none", "combined_relational_operations_require_calibration" ]	structural_heuristic	Qwen3.6-27B	heuristic_not_model_measured
gretel_sql:4aa73a02c9655084807c496d82864ca7f306db6c50f2c66817d67a263aa2e261	task:5cc93490e8a9c5ef0211c436d805090ef927f4a7dc93663bf8ff93a987de2e9b	0f5c27e213b43e19c4d48df757464d6cffefce0212a2dd897097aa5e642a813f	gretel_sql	gretelai/synthetic_text_to_sql	740ab236e64503fba51be1101df7a1be83bf455d	synthetic_text_to_sql_test.snappy.parquet	109	test	https://huggingface.co/datasets/gretelai/synthetic_text_to_sql/resolve/740ab236e64503fba51be1101df7a1be83bf455d/synthetic_text_to_sql_test.snappy.parquet	apache-2.0	Gretel Synthetic Text-to-SQL dataset creators; source card and author citation preserved in the download manifest.	proposed	Which defense contracts have the highest total value, and what are their respective values?	CREATE TABLE Defense_Contracts (Contract_ID INT, Contract_Name VARCHAR(255), Agency VARCHAR(255), Value DECIMAL(18,2)); INSERT INTO Defense_Contracts (Contract_ID, Contract_Name, Agency, Value) VALUES (1, 'Contract A', 'DOD', 5000000), (2, 'Contract B', 'DOJ', 6000000), (3, 'Contract C', 'DOD', 7000000), (4, 'Contract ...	database_business_operations	database_business_operations.database_query_analysis	heuristic_unreviewed	Mapped from source sql_task_type='analytics and reporting'; no individual SQL execution or semantic review.	en	none	SELECT Contract_Name, Value FROM (SELECT Contract_Name, Value, ROW_NUMBER() OVER (ORDER BY Value DESC) as Rank FROM Defense_Contracts) as Ranked_Contracts WHERE Rank <= 3;	{"domain":"defense industry","domain_description":"Defense contract data, military equipment maintenance, threat intelligence metrics, and veteran employment stats.","id":109,"sql_complexity":"window functions","sql_complexity_description":"window functions (e.g., ROW_NUMBER, LEAD, LAG, RANk, NTILE, PERCENT_RANK, etc.)...	f9fcd425b721e4863a77910f8524b34f3e16da0574d8dc2efdb021f9953b69e2	gretel_sql:test:109	window functions	intermediate	Estimated from source sql_complexity='window functions'; query feature labels are only proxies for difficulty, not measured solver performance. No empirical solver calibration was performed.	heuristic	needs_difficulty_calibration	not_measured	analysis_only_unvalidated	primary	task_annotation	108	[ "difficulty_not_calibrated", "not_individually_reviewed", "reference_not_independently_verified", "reference_or_constraint_only_no_rubric", "response_diversity_not_validated", "source_evaluation_split", "sql_dialect_and_result_equivalence_need_review", "sql_not_executed", "synthetic_reference" ]	false	true	false	task:5cc93490e8a9c5ef0211c436d805090ef927f4a7dc93663bf8ff93a987de2e9b	1	1	[ "gretel_sql" ]	held_out	curated_candidate	[]	core_candidate	[ "source_sql_complexity:window functions", "sql_structure:selects=2;joins=0;statements=1;advanced=OVER", "combined_relational_operations_require_calibration" ]	structural_heuristic	Qwen3.6-27B	heuristic_not_model_measured
gretel_sql:ac5f0ae715b67600dbf8ce5a233b447ec91698ea080464b3217e0c407055e66d	task:a72b7dc3bb4ac92b6d07577365bf38326217f46518a2b52fe09248e5c50dcd1f	67988782d10d9406a2a2393df036571e86c130af3fe9ee728da96a9a7404b688	gretel_sql	gretelai/synthetic_text_to_sql	740ab236e64503fba51be1101df7a1be83bf455d	synthetic_text_to_sql_test.snappy.parquet	114	test	https://huggingface.co/datasets/gretelai/synthetic_text_to_sql/resolve/740ab236e64503fba51be1101df7a1be83bf455d/synthetic_text_to_sql_test.snappy.parquet	apache-2.0	Gretel Synthetic Text-to-SQL dataset creators; source card and author citation preserved in the download manifest.	proposed	What is the average weight of cargo handled by vessels in the 'Bulk Carrier' type at each port?	CREATE TABLE ports (id INT, name VARCHAR(50), location VARCHAR(50), un_code VARCHAR(10)); CREATE TABLE vessels (id INT, name VARCHAR(50), type VARCHAR(50), year_built INT, port_id INT); CREATE TABLE cargo (id INT, description VARCHAR(50), weight FLOAT, port_id INT, vessel_id INT); CREATE VIEW vessel_cargo AS SELECT v.n...	database_business_operations	database_business_operations.database_query_analysis	heuristic_unreviewed	Mapped from source sql_task_type='analytics and reporting'; no individual SQL execution or semantic review.	en	none	SELECT p.name AS port_name, AVG(vc.weight) AS avg_weight FROM ports p JOIN vessels v ON p.id = v.port_id JOIN vessel_cargo vc ON v.name = vc.vessel_name WHERE v.type = 'Bulk Carrier' GROUP BY p.name;	{"domain":"ocean shipping","domain_description":"Detailed records on cargo handling, fleet management, port operations, and regulatory compliance in ocean shipping.","id":114,"sql_complexity":"multiple_joins","sql_complexity_description":"two or more joins (specify inner, outer, cross)","sql_explanation":"This query ca...	cbc90ad494e107f7e99a3250f4fac61ed8b7cec945d93ac89348c6bfce39a51d	gretel_sql:test:114	multiple_joins	intermediate	Estimated from source sql_complexity='multiple_joins'; query feature labels are only proxies for difficulty, not measured solver performance. No empirical solver calibration was performed.	heuristic	needs_difficulty_calibration	not_measured	analysis_only_unvalidated	primary	task_annotation	113	[ "difficulty_not_calibrated", "not_individually_reviewed", "reference_not_independently_verified", "reference_or_constraint_only_no_rubric", "response_diversity_not_validated", "source_evaluation_split", "sql_dialect_and_result_equivalence_need_review", "sql_not_executed", "synthetic_reference" ]	false	true	false	task:a72b7dc3bb4ac92b6d07577365bf38326217f46518a2b52fe09248e5c50dcd1f	1	1	[ "gretel_sql" ]	held_out	curated_candidate	[]	core_candidate	[ "source_sql_complexity:multiple_joins", "sql_structure:selects=1;joins=2;statements=1;advanced=none", "combined_relational_operations_require_calibration" ]	structural_heuristic	Qwen3.6-27B	heuristic_not_model_measured
gretel_sql:5a3983d437ea7313659783169e052c2fcc99ee5002db9768a5d4e16f9aacba4d	task:668d9ed3342193821fcf71aa112035d491236e82ba7a3a323df7ed6b554e99ae	ac036d0a37566ef33e478c11e478c09f9035bdb571b0abd7403dff2cf3cbb4a7	gretel_sql	gretelai/synthetic_text_to_sql	740ab236e64503fba51be1101df7a1be83bf455d	synthetic_text_to_sql_test.snappy.parquet	221	test	https://huggingface.co/datasets/gretelai/synthetic_text_to_sql/resolve/740ab236e64503fba51be1101df7a1be83bf455d/synthetic_text_to_sql_test.snappy.parquet	apache-2.0	Gretel Synthetic Text-to-SQL dataset creators; source card and author citation preserved in the download manifest.	proposed	Which mental health conditions were treated most frequently in Canada during 2022?	CREATE TABLE patients (id INT, country VARCHAR(255)); CREATE TABLE treatments (id INT, patient_id INT, treatment_date DATE); CREATE TABLE conditions (id INT, patient_id INT, condition VARCHAR(255)); INSERT INTO patients (id, country) VALUES (1, 'Canada'), (2, 'Canada'); INSERT INTO treatments (id, patient_id, treatment...	database_business_operations	database_business_operations.database_query_analysis	heuristic_unreviewed	Mapped from source sql_task_type='analytics and reporting'; no individual SQL execution or semantic review.	en	none	SELECT conditions.condition, COUNT(conditions.condition) AS count FROM conditions JOIN patients ON conditions.patient_id = patients.id JOIN treatments ON patients.id = treatments.patient_id WHERE patients.country = 'Canada' AND treatments.treatment_date >= '2022-01-01' AND treatments.treatment_date < '2023-01-01' GROUP...	{"domain":"mental health","domain_description":"In-depth data on mental health conditions, treatment approaches, patient outcomes, and public awareness campaigns in mental health.","id":221,"sql_complexity":"multiple_joins","sql_complexity_description":"two or more joins (specify inner, outer, cross)","sql_explanation"...	1ade5c12ba44207339a049f5bedfe5c77f6a4b6cf22d53ace16a487a72389f68	gretel_sql:test:221	multiple_joins	intermediate	Estimated from source sql_complexity='multiple_joins'; query feature labels are only proxies for difficulty, not measured solver performance. No empirical solver calibration was performed.	heuristic	needs_difficulty_calibration	not_measured	analysis_only_unvalidated	primary	task_annotation	220	[ "difficulty_not_calibrated", "not_individually_reviewed", "reference_not_independently_verified", "reference_or_constraint_only_no_rubric", "response_diversity_not_validated", "source_evaluation_split", "sql_dialect_and_result_equivalence_need_review", "sql_not_executed", "synthetic_reference" ]	false	true	false	task:668d9ed3342193821fcf71aa112035d491236e82ba7a3a323df7ed6b554e99ae	1	1	[ "gretel_sql" ]	held_out	curated_candidate	[]	core_candidate	[ "source_sql_complexity:multiple_joins", "sql_structure:selects=1;joins=2;statements=1;advanced=none", "combined_relational_operations_require_calibration" ]	structural_heuristic	Qwen3.6-27B	heuristic_not_model_measured
gretel_sql:7e20e548f8a175e0b0b2a6519f174bb728f2f13d78123e4bb5365200d2eae019	task:daf495e597b052523601e658cddeb5b2c335fb3edce926095df46755a04cf665	be005e0c93da36ce0eab4e13780dbef8eb25b6bf0a3c5fbdd4f0cd4e6759c093	gretel_sql	gretelai/synthetic_text_to_sql	740ab236e64503fba51be1101df7a1be83bf455d	synthetic_text_to_sql_test.snappy.parquet	230	test	https://huggingface.co/datasets/gretelai/synthetic_text_to_sql/resolve/740ab236e64503fba51be1101df7a1be83bf455d/synthetic_text_to_sql_test.snappy.parquet	apache-2.0	Gretel Synthetic Text-to-SQL dataset creators; source card and author citation preserved in the download manifest.	proposed	Find the mobile subscribers with consecutive speed drops greater than 25% for the last 3 months, ordered by subscription IDs.	CREATE TABLE mobile_usage_detailed (subscriber_id INT, month INT, speed FLOAT); INSERT INTO mobile_usage_detailed (subscriber_id, month, speed) VALUES (1, 1, 100), (1, 2, 80), (1, 3, 70), (2, 1, 200), (2, 2, 180), (2, 3, 160), (3, 1, 150), (3, 2, 130), (3, 3, 110);	database_business_operations	database_business_operations.database_query_analysis	heuristic_unreviewed	Mapped from source sql_task_type='analytics and reporting'; no individual SQL execution or semantic review.	en	none	SELECT subscriber_id, speed, month FROM (SELECT subscriber_id, speed, month, LAG(speed, 1) OVER (PARTITION BY subscriber_id ORDER BY month) as prev_speed, LAG(speed, 2) OVER (PARTITION BY subscriber_id ORDER BY month) as prev_prev_speed FROM mobile_usage_detailed) t WHERE t.speed < 0.75 * t.prev_speed AND t.speed < 0.7...	{"domain":"telecommunications","domain_description":"Mobile and broadband subscriber data, network infrastructure investments, customer usage patterns, and regulatory compliance information.","id":230,"sql_complexity":"window functions","sql_complexity_description":"window functions (e.g., ROW_NUMBER, LEAD, LAG, RANk, ...	112b3a088b4574579d30ffc11195b74bc894217dda47629d876a18ec9141a0bb	gretel_sql:test:230	window functions	intermediate	Estimated from source sql_complexity='window functions'; query feature labels are only proxies for difficulty, not measured solver performance. No empirical solver calibration was performed.	heuristic	needs_difficulty_calibration	not_measured	analysis_only_unvalidated	primary	task_annotation	229	[ "difficulty_not_calibrated", "not_individually_reviewed", "reference_not_independently_verified", "reference_or_constraint_only_no_rubric", "response_diversity_not_validated", "source_evaluation_split", "sql_dialect_and_result_equivalence_need_review", "sql_not_executed", "synthetic_reference" ]	false	true	false	task:daf495e597b052523601e658cddeb5b2c335fb3edce926095df46755a04cf665	1	1	[ "gretel_sql" ]	held_out	curated_candidate	[]	core_candidate	[ "source_sql_complexity:window functions", "sql_structure:selects=2;joins=0;statements=1;advanced=OVER", "combined_relational_operations_require_calibration" ]	structural_heuristic	Qwen3.6-27B	heuristic_not_model_measured
gretel_sql:6bb8af4984d6034a3e9b32ff1346042eb58c39ad879aa32c4a6f5709edd08185	task:a3871b1db697742e405cdd20bd7f55ca3f34b6bf88d9b3c67c38b2a05b9d1d6b	8f64fc8cfb445395c52a2b3f77859683574cead87f8b1cb12dc18c687671528a	gretel_sql	gretelai/synthetic_text_to_sql	740ab236e64503fba51be1101df7a1be83bf455d	synthetic_text_to_sql_test.snappy.parquet	520	test	https://huggingface.co/datasets/gretelai/synthetic_text_to_sql/resolve/740ab236e64503fba51be1101df7a1be83bf455d/synthetic_text_to_sql_test.snappy.parquet	apache-2.0	Gretel Synthetic Text-to-SQL dataset creators; source card and author citation preserved in the download manifest.	proposed	What is the total number of electric vehicles sold in 'California' in the 'sales' schema?	CREATE TABLE sales_regions (id INT, name VARCHAR(50)); CREATE TABLE sales (id INT, region_id INT, vehicle_count INT); CREATE TABLE vehicles (id INT, type VARCHAR(50)); INSERT INTO sales_regions VALUES (1, 'California'); INSERT INTO sales VALUES (1, 1, 5000); INSERT INTO vehicles VALUES (1, 'electric');	database_business_operations	database_business_operations.database_query_analysis	heuristic_unreviewed	Mapped from source sql_task_type='analytics and reporting'; no individual SQL execution or semantic review.	en	none	SELECT SUM(sales.vehicle_count) FROM sales INNER JOIN sales_regions ON sales.region_id = sales_regions.id INNER JOIN vehicles ON sales.id = vehicles.id WHERE vehicles.type = 'electric' AND sales_regions.name = 'California';	{"domain":"automotive","domain_description":"Vehicle safety testing results, autonomous driving research data, electric vehicle adoption statistics, and auto show information.","id":520,"sql_complexity":"multiple_joins","sql_complexity_description":"two or more joins (specify inner, outer, cross)","sql_explanation":"Fi...	9d38bf9cb28f33e9e5ea04e6a96c8e8d545757cf665a2ed0abde4792de84157a	gretel_sql:test:520	multiple_joins	intermediate	Estimated from source sql_complexity='multiple_joins'; query feature labels are only proxies for difficulty, not measured solver performance. No empirical solver calibration was performed.	heuristic	needs_difficulty_calibration	not_measured	analysis_only_unvalidated	primary	task_annotation	519	[ "difficulty_not_calibrated", "not_individually_reviewed", "reference_not_independently_verified", "reference_or_constraint_only_no_rubric", "response_diversity_not_validated", "source_evaluation_split", "sql_dialect_and_result_equivalence_need_review", "sql_not_executed", "synthetic_reference" ]	false	true	false	task:a3871b1db697742e405cdd20bd7f55ca3f34b6bf88d9b3c67c38b2a05b9d1d6b	1	1	[ "gretel_sql" ]	held_out	curated_candidate	[]	core_candidate	[ "source_sql_complexity:multiple_joins", "sql_structure:selects=1;joins=2;statements=1;advanced=none", "combined_relational_operations_require_calibration" ]	structural_heuristic	Qwen3.6-27B	heuristic_not_model_measured
gretel_sql:87c4e12fe199ca347bf97c7e00082e4b70680965bef34d32f9654ae7ea2b33e9	task:651a9a5de76a2827874f63c3aa44cfac2c2b54f44eda7c0fdf9ca8ecfd4e2bb8	bab44501738294e19a13688cbababae07971f41a087ec3d4607480fcb6d33ab4	gretel_sql	gretelai/synthetic_text_to_sql	740ab236e64503fba51be1101df7a1be83bf455d	synthetic_text_to_sql_test.snappy.parquet	526	test	https://huggingface.co/datasets/gretelai/synthetic_text_to_sql/resolve/740ab236e64503fba51be1101df7a1be83bf455d/synthetic_text_to_sql_test.snappy.parquet	apache-2.0	Gretel Synthetic Text-to-SQL dataset creators; source card and author citation preserved in the download manifest.	proposed	What is the percentage of accidents for each aircraft model?	CREATE SCHEMA if not exists aerospace;CREATE TABLE if not exists aerospace.aircraft (id INT PRIMARY KEY, name VARCHAR(50), model VARCHAR(50), accidents INT); INSERT INTO aerospace.aircraft (id, name, model, accidents) VALUES (1, 'Boeing', '737', 3), (2, 'Boeing', '747', 2), (3, 'Airbus', 'A320', 6);	database_business_operations	database_business_operations.database_query_analysis	heuristic_unreviewed	Mapped from source sql_task_type='analytics and reporting'; no individual SQL execution or semantic review.	en	none	SELECT model, (SUM(accidents) OVER (PARTITION BY model) * 100.0 / (SELECT SUM(accidents) FROM aerospace.aircraft)) as accident_percentage FROM aerospace.aircraft;	{"domain":"aerospace","domain_description":"Aircraft manufacturing data, satellite deployment projects, flight safety records, and space exploration research.","id":526,"sql_complexity":"window functions","sql_complexity_description":"window functions (e.g., ROW_NUMBER, LEAD, LAG, RANk, NTILE, PERCENT_RANK, etc.) with ...	38ab81904e9d975f69d0fa05ca547db4015474ec300712b5d55336f36abcbaf9	gretel_sql:test:526	window functions	intermediate	Estimated from source sql_complexity='window functions'; query feature labels are only proxies for difficulty, not measured solver performance. No empirical solver calibration was performed.	heuristic	needs_difficulty_calibration	not_measured	analysis_only_unvalidated	primary	task_annotation	525	[ "difficulty_not_calibrated", "not_individually_reviewed", "reference_not_independently_verified", "reference_or_constraint_only_no_rubric", "response_diversity_not_validated", "source_evaluation_split", "sql_dialect_and_result_equivalence_need_review", "sql_not_executed", "synthetic_reference" ]	false	true	false	task:651a9a5de76a2827874f63c3aa44cfac2c2b54f44eda7c0fdf9ca8ecfd4e2bb8	1	1	[ "gretel_sql" ]	held_out	curated_candidate	[]	core_candidate	[ "source_sql_complexity:window functions", "sql_structure:selects=2;joins=0;statements=1;advanced=OVER", "combined_relational_operations_require_calibration" ]	structural_heuristic	Qwen3.6-27B	heuristic_not_model_measured
gretel_sql:65cdf87508384e3bd00b6611e3293688bece5d4dcdc517cdb3d8539acd600c60	task:cc75dc0b60e865f0c5430eacb08019bed52c3ef6a60b453a0b7dd5856be3e89a	199f90e6943fe6c9fb106cf606f6027150300ef25d952cf88f84aff058ad05cc	gretel_sql	gretelai/synthetic_text_to_sql	740ab236e64503fba51be1101df7a1be83bf455d	synthetic_text_to_sql_test.snappy.parquet	530	test	https://huggingface.co/datasets/gretelai/synthetic_text_to_sql/resolve/740ab236e64503fba51be1101df7a1be83bf455d/synthetic_text_to_sql_test.snappy.parquet	apache-2.0	Gretel Synthetic Text-to-SQL dataset creators; source card and author citation preserved in the download manifest.	proposed	What is the maximum feeding rate by feed type and farm size?	CREATE TABLE Feed ( id INT PRIMARY KEY, type VARCHAR(50) ); CREATE TABLE Farm ( id INT PRIMARY KEY, feed_id INT, size INT, FOREIGN KEY (feed_id) REFERENCES Feed(id) ); CREATE TABLE FeedingRate ( farm_id INT, feed_id INT, rate INT, FOREIGN KEY (farm_id) REFERENCES Farm(id), FOREIGN KEY (feed_id) REFERENCES Feed(id) );	database_business_operations	database_business_operations.database_query_analysis	heuristic_unreviewed	Mapped from source sql_task_type='analytics and reporting'; no individual SQL execution or semantic review.	en	none	SELECT Feed.type, Farm.size, MAX(FeedingRate.rate) FROM Feed INNER JOIN FeedingRate ON Feed.id = FeedingRate.feed_id INNER JOIN Farm ON FeedingRate.farm_id = Farm.id GROUP BY Feed.type, Farm.size;	{"domain":"aquaculture","domain_description":"Aquatic farming data, fish stock management, ocean health metrics, and sustainable seafood trends.","id":530,"sql_complexity":"multiple_joins","sql_complexity_description":"two or more joins (specify inner, outer, cross)","sql_explanation":"The SQL query joins the Feed, Fee...	1292890e22ff60e64aef70f7b4b88f852ef382ddbd049c8d67331dc343a48975	gretel_sql:test:530	multiple_joins	intermediate	Estimated from source sql_complexity='multiple_joins'; query feature labels are only proxies for difficulty, not measured solver performance. No empirical solver calibration was performed.	heuristic	needs_difficulty_calibration	not_measured	analysis_only_unvalidated	primary	task_annotation	529	[ "difficulty_not_calibrated", "not_individually_reviewed", "reference_not_independently_verified", "reference_or_constraint_only_no_rubric", "response_diversity_not_validated", "source_evaluation_split", "sql_dialect_and_result_equivalence_need_review", "sql_not_executed", "synthetic_reference" ]	false	true	false	task:cc75dc0b60e865f0c5430eacb08019bed52c3ef6a60b453a0b7dd5856be3e89a	1	1	[ "gretel_sql" ]	held_out	curated_candidate	[]	core_candidate	[ "source_sql_complexity:multiple_joins", "sql_structure:selects=1;joins=2;statements=1;advanced=none", "combined_relational_operations_require_calibration" ]	structural_heuristic	Qwen3.6-27B	heuristic_not_model_measured
gretel_sql:638fa07b6e6620e4ae8c7114ef15a6867b63cbb5fe0ba84ac8a902973e24397c	task:f0e68e38897af2d9b4357b3e6d6d88803b3a6b4a358a52c7947517ca280b9bd8	f8fdfd74643d42004277a302f3d71f5a22dc70116726020732f517f2947af1dc	gretel_sql	gretelai/synthetic_text_to_sql	740ab236e64503fba51be1101df7a1be83bf455d	synthetic_text_to_sql_test.snappy.parquet	558	test	https://huggingface.co/datasets/gretelai/synthetic_text_to_sql/resolve/740ab236e64503fba51be1101df7a1be83bf455d/synthetic_text_to_sql_test.snappy.parquet	apache-2.0	Gretel Synthetic Text-to-SQL dataset creators; source card and author citation preserved in the download manifest.	proposed	How many professional development courses were completed by teachers in the English department?	CREATE TABLE teachers (teacher_id INT, department_id INT, teacher_name VARCHAR(255)); INSERT INTO teachers VALUES (1, 1, 'Ms. Hernandez'); INSERT INTO teachers VALUES (2, 2, 'Mr. Johnson'); CREATE TABLE departments (department_id INT, department_name VARCHAR(255)); INSERT INTO departments VALUES (1, 'English'); INSERT ...	database_business_operations	database_business_operations.database_query_analysis	heuristic_unreviewed	Mapped from source sql_task_type='analytics and reporting'; no individual SQL execution or semantic review.	en	none	SELECT d.department_name, COUNT(c.course_id) FROM course_enrollment ce INNER JOIN teachers t ON ce.teacher_id = t.teacher_id INNER JOIN departments d ON t.department_id = d.department_id WHERE d.department_name = 'English';	{"domain":"education","domain_description":"Education data on student mental health, teacher professional development, open pedagogy, and lifelong learning.","id":558,"sql_complexity":"multiple_joins","sql_complexity_description":"two or more joins (specify inner, outer, cross)","sql_explanation":"This query performs a...	68e560f79fddabea2843cdd27372a25eb05682d3b88e8e22eab27cba4bb6672c	gretel_sql:test:558	multiple_joins	intermediate	Estimated from source sql_complexity='multiple_joins'; query feature labels are only proxies for difficulty, not measured solver performance. No empirical solver calibration was performed.	heuristic	needs_difficulty_calibration	not_measured	analysis_only_unvalidated	primary	task_annotation	557	[ "difficulty_not_calibrated", "not_individually_reviewed", "reference_not_independently_verified", "reference_or_constraint_only_no_rubric", "response_diversity_not_validated", "source_evaluation_split", "sql_dialect_and_result_equivalence_need_review", "sql_not_executed", "synthetic_reference" ]	false	true	false	task:f0e68e38897af2d9b4357b3e6d6d88803b3a6b4a358a52c7947517ca280b9bd8	1	1	[ "gretel_sql" ]	held_out	curated_candidate	[]	core_candidate	[ "source_sql_complexity:multiple_joins", "sql_structure:selects=1;joins=2;statements=1;advanced=none", "combined_relational_operations_require_calibration" ]	structural_heuristic	Qwen3.6-27B	heuristic_not_model_measured
gretel_sql:605eb1151f2bd663d78aa6d8914e4f1bbb6b3b94a4533f363f26b8eaf9a26853	task:a1139ac892b43b22dcc4fcff41c14fdea6e7228520f879f048bed06ff6012a3c	2341f0102a8336a0623162f960aeb4ff5f3b392216fee8894e6abc4f16e45828	gretel_sql	gretelai/synthetic_text_to_sql	740ab236e64503fba51be1101df7a1be83bf455d	synthetic_text_to_sql_test.snappy.parquet	678	test	https://huggingface.co/datasets/gretelai/synthetic_text_to_sql/resolve/740ab236e64503fba51be1101df7a1be83bf455d/synthetic_text_to_sql_test.snappy.parquet	apache-2.0	Gretel Synthetic Text-to-SQL dataset creators; source card and author citation preserved in the download manifest.	proposed	What is the average labor productivity by mine type in the past 12 months?	CREATE TABLE mine_labor_productivity (mine_type VARCHAR(255), productivity NUMERIC, measurement_date DATE); INSERT INTO mine_labor_productivity (mine_type, productivity, measurement_date) VALUES ('open_pit', 1234, '2021-08-01'), ('underground', 2345, '2021-08-01'), ('open_pit', 5432, '2021-07-01'), ('underground', 6789...	database_business_operations	database_business_operations.database_query_analysis	heuristic_unreviewed	Mapped from source sql_task_type='analytics and reporting'; no individual SQL execution or semantic review.	en	none	SELECT mine_type, AVG(productivity) as avg_productivity FROM (SELECT mine_type, productivity, measurement_date, ROW_NUMBER() OVER (PARTITION BY mine_type ORDER BY measurement_date DESC) as rn FROM mine_labor_productivity WHERE measurement_date >= DATEADD(month, -12, CURRENT_DATE)) t WHERE rn = 1 GROUP BY mine_type;	{"domain":"mining industry","domain_description":"Mineral extraction statistics, environmental impact assessments, labor productivity metrics, and geological survey information.","id":678,"sql_complexity":"window functions","sql_complexity_description":"window functions (e.g., ROW_NUMBER, LEAD, LAG, RANk, NTILE, PERCEN...	587654e1d2bdf78b465061da0cf60f6d43f8b83a087923e2433c74c6335dd05a	gretel_sql:test:678	window functions	intermediate	Estimated from source sql_complexity='window functions'; query feature labels are only proxies for difficulty, not measured solver performance. No empirical solver calibration was performed.	heuristic	needs_difficulty_calibration	not_measured	analysis_only_unvalidated	primary	task_annotation	677	[ "difficulty_not_calibrated", "not_individually_reviewed", "reference_not_independently_verified", "reference_or_constraint_only_no_rubric", "response_diversity_not_validated", "source_evaluation_split", "sql_dialect_and_result_equivalence_need_review", "sql_not_executed", "synthetic_reference" ]	false	true	false	task:a1139ac892b43b22dcc4fcff41c14fdea6e7228520f879f048bed06ff6012a3c	1	1	[ "gretel_sql" ]	held_out	curated_candidate	[]	core_candidate	[ "source_sql_complexity:window functions", "sql_structure:selects=2;joins=0;statements=1;advanced=OVER", "combined_relational_operations_require_calibration" ]	structural_heuristic	Qwen3.6-27B	heuristic_not_model_measured
gretel_sql:1140fe7186b632d58ed096be6fcc60542e87a5d06d81be0050471bc91eeb2b54	task:9ba448bbb897e7afcf0455f3aecc67154b6af21bdab77fb862718386fe8a75db	c5ccf013a7ed323d8af9f17fadd52a465d3eda3b76e42c9e20fc0ab35820f0e2	gretel_sql	gretelai/synthetic_text_to_sql	740ab236e64503fba51be1101df7a1be83bf455d	synthetic_text_to_sql_test.snappy.parquet	682	test	https://huggingface.co/datasets/gretelai/synthetic_text_to_sql/resolve/740ab236e64503fba51be1101df7a1be83bf455d/synthetic_text_to_sql_test.snappy.parquet	apache-2.0	Gretel Synthetic Text-to-SQL dataset creators; source card and author citation preserved in the download manifest.	proposed	What is the total number of articles written by each author in each region?	CREATE TABLE authors (id INT, name TEXT); CREATE TABLE regions (id INT, name TEXT); CREATE TABLE articles (id INT, title TEXT, content TEXT, author_id INT, region_id INT); INSERT INTO authors (id, name) VALUES (1, 'John Doe'), (2, 'Jane Smith'); INSERT INTO regions (id, name) VALUES (1, 'North'), (2, 'South'), (3, 'Eas...	database_business_operations	database_business_operations.database_query_analysis	heuristic_unreviewed	Mapped from source sql_task_type='analytics and reporting'; no individual SQL execution or semantic review.	en	none	SELECT authors.name, regions.name, COUNT(articles.id) FROM authors INNER JOIN articles ON authors.id = articles.author_id INNER JOIN regions ON regions.id = articles.region_id GROUP BY authors.name, regions.name;	{"domain":"journalism","domain_description":"News reporting data, audience demographics, media ethics information, and investigative journalism projects.","id":682,"sql_complexity":"multiple_joins","sql_complexity_description":"two or more joins (specify inner, outer, cross)","sql_explanation":"This SQL query retrieves...	595530aa682c3b5c863053849e0c392fab3ca7cf706fce9f209e6ed737c60080	gretel_sql:test:682	multiple_joins	intermediate	Estimated from source sql_complexity='multiple_joins'; query feature labels are only proxies for difficulty, not measured solver performance. No empirical solver calibration was performed.	heuristic	needs_difficulty_calibration	not_measured	analysis_only_unvalidated	primary	task_annotation	681	[ "difficulty_not_calibrated", "not_individually_reviewed", "reference_not_independently_verified", "reference_or_constraint_only_no_rubric", "response_diversity_not_validated", "source_evaluation_split", "sql_dialect_and_result_equivalence_need_review", "sql_not_executed", "synthetic_reference" ]	false	true	false	task:9ba448bbb897e7afcf0455f3aecc67154b6af21bdab77fb862718386fe8a75db	1	1	[ "gretel_sql" ]	held_out	curated_candidate	[]	core_candidate	[ "source_sql_complexity:multiple_joins", "sql_structure:selects=1;joins=2;statements=1;advanced=none", "combined_relational_operations_require_calibration" ]	structural_heuristic	Qwen3.6-27B	heuristic_not_model_measured
gretel_sql:2cb817c0878d4ceca8e51cbe2251ddaab269ba243af0133f8df8b97c10542591	task:9816b122d966001a0e963d05552c2c81c94d9e51bff1cfcb3eaae574f60afeaf	9703fe06ce5ef646effd3caba6dbb2eeffc909cdec5569e5f71325d866cc4eb3	gretel_sql	gretelai/synthetic_text_to_sql	740ab236e64503fba51be1101df7a1be83bf455d	synthetic_text_to_sql_test.snappy.parquet	735	test	https://huggingface.co/datasets/gretelai/synthetic_text_to_sql/resolve/740ab236e64503fba51be1101df7a1be83bf455d/synthetic_text_to_sql_test.snappy.parquet	apache-2.0	Gretel Synthetic Text-to-SQL dataset creators; source card and author citation preserved in the download manifest.	proposed	Show the total assets under management (AUM) for each investment strategy.	CREATE TABLE investment_strategies (strategy_id INT, strategy VARCHAR(20)); INSERT INTO investment_strategies (strategy_id, strategy) VALUES (1, 'Conservative'), (2, 'Moderate'), (3, 'Aggressive'); CREATE TABLE client_strategy (client_id INT, strategy_id INT); INSERT INTO client_strategy (client_id, strategy_id) VALUES...	database_business_operations	database_business_operations.database_query_analysis	heuristic_unreviewed	Mapped from source sql_task_type='analytics and reporting'; no individual SQL execution or semantic review.	en	none	SELECT cs.strategy, SUM(value) AS total_aum FROM client_strategy cs JOIN clients c ON cs.client_id = c.client_id JOIN assets a ON c.client_id = a.client_id GROUP BY cs.strategy;	{"domain":"financial services","domain_description":"Detailed financial data including investment strategies, risk management, fraud detection, customer analytics, and regulatory compliance.","id":735,"sql_complexity":"multiple_joins","sql_complexity_description":"two or more joins (specify inner, outer, cross)","sql_e...	c69a1cb177a81cee54cae23948e51d766b1dbc072cb1187ddbecb3bb3a8a5986	gretel_sql:test:735	multiple_joins	intermediate	Estimated from source sql_complexity='multiple_joins'; query feature labels are only proxies for difficulty, not measured solver performance. No empirical solver calibration was performed.	heuristic	needs_difficulty_calibration	not_measured	analysis_only_unvalidated	primary	task_annotation	734	[ "difficulty_not_calibrated", "not_individually_reviewed", "reference_not_independently_verified", "reference_or_constraint_only_no_rubric", "response_diversity_not_validated", "source_evaluation_split", "sql_dialect_and_result_equivalence_need_review", "sql_not_executed", "synthetic_reference" ]	false	true	false	task:9816b122d966001a0e963d05552c2c81c94d9e51bff1cfcb3eaae574f60afeaf	1	1	[ "gretel_sql" ]	held_out	curated_candidate	[]	core_candidate	[ "source_sql_complexity:multiple_joins", "sql_structure:selects=1;joins=2;statements=1;advanced=none", "combined_relational_operations_require_calibration" ]	structural_heuristic	Qwen3.6-27B	heuristic_not_model_measured
gretel_sql:7459d61c01da813f6ad9cbb9b4229d0cf7ba56e8dfa410d8e4a34583baa08390	task:70f33e41fbcef91a27881f142726199b85e75bbb40417c18833e85b72ba12a31	aaa07784f839ac7d7cfb0da7045e5af9beb07288fed3355722086d4b68ffae18	gretel_sql	gretelai/synthetic_text_to_sql	740ab236e64503fba51be1101df7a1be83bf455d	synthetic_text_to_sql_test.snappy.parquet	746	test	https://huggingface.co/datasets/gretelai/synthetic_text_to_sql/resolve/740ab236e64503fba51be1101df7a1be83bf455d/synthetic_text_to_sql_test.snappy.parquet	apache-2.0	Gretel Synthetic Text-to-SQL dataset creators; source card and author citation preserved in the download manifest.	proposed	Who has the highest number of wins as a coach for each team in a season?	CREATE TABLE Coach (CoachID int, CoachName varchar(50), TeamID int); CREATE TABLE Match (MatchID int, HomeTeamID int, AwayTeamID int, HomeTeamResult varchar(5)); INSERT INTO Coach (CoachID, CoachName, TeamID) VALUES (1, 'Jose Mourinho', 1), (2, 'Pep Guardiola', 1), (3, 'Jurgen Klopp', 2), (4, 'Mauricio Pochettino', 2)...	database_business_operations	database_business_operations.database_query_analysis	heuristic_unreviewed	Mapped from source sql_task_type='analytics and reporting'; no individual SQL execution or semantic review.	en	none	SELECT c.TeamID, c.CoachName, COUNT(CASE WHEN h.HomeTeamResult = 'Win' THEN 1 END) + COUNT(CASE WHEN a.HomeTeamResult = 'Win' THEN 1 END) AS Wins FROM Coach c LEFT JOIN Match h ON c.TeamID = h.HomeTeamID AND h.HomeTeamResult = 'Win' LEFT JOIN Match a ON c.TeamID = a.AwayTeamID AND a.HomeTeamResult = 'Win' GROUP BY c.Te...	{"domain":"sports","domain_description":"Extensive data on athlete performance, team management, fan engagement, facility operations, and event planning in sports.","id":746,"sql_complexity":"multiple_joins","sql_complexity_description":"two or more joins (specify inner, outer, cross)","sql_explanation":"We calculate t...	63fc7c03b5b1ef1e6264594bd37f803866784b1511508639b707998eea5cb854	gretel_sql:test:746	multiple_joins	intermediate	Estimated from source sql_complexity='multiple_joins'; query feature labels are only proxies for difficulty, not measured solver performance. No empirical solver calibration was performed.	heuristic	needs_difficulty_calibration	not_measured	analysis_only_unvalidated	primary	task_annotation	745	[ "difficulty_not_calibrated", "not_individually_reviewed", "reference_not_independently_verified", "reference_or_constraint_only_no_rubric", "response_diversity_not_validated", "source_evaluation_split", "sql_dialect_and_result_equivalence_need_review", "sql_not_executed", "synthetic_reference" ]	false	true	false	task:70f33e41fbcef91a27881f142726199b85e75bbb40417c18833e85b72ba12a31	1	1	[ "gretel_sql" ]	held_out	curated_candidate	[]	core_candidate	[ "source_sql_complexity:multiple_joins", "sql_structure:selects=1;joins=2;statements=1;advanced=none", "combined_relational_operations_require_calibration" ]	structural_heuristic	Qwen3.6-27B	heuristic_not_model_measured
gretel_sql:01d48c68838101488159909639607f2b3ef488591cb373e254b64b2df4aa9c31	task:bc95ae27ab56e584f96e47115c558f8119bd5a8b666d2592292d0451cbaa4462	6063ff0c84448751a65d95d66772e00888e5ae9eda1e539a8d266a962c5728f8	gretel_sql	gretelai/synthetic_text_to_sql	740ab236e64503fba51be1101df7a1be83bf455d	synthetic_text_to_sql_test.snappy.parquet	762	test	https://huggingface.co/datasets/gretelai/synthetic_text_to_sql/resolve/740ab236e64503fba51be1101df7a1be83bf455d/synthetic_text_to_sql_test.snappy.parquet	apache-2.0	Gretel Synthetic Text-to-SQL dataset creators; source card and author citation preserved in the download manifest.	proposed	What is the name and location of the top 3 most donated heritage sites?	CREATE TABLE HeritageSites (SiteID int, Name varchar(100), Location varchar(100), TotalDonations decimal(10,2)); INSERT INTO HeritageSites (SiteID, Name, Location, TotalDonations) VALUES (1, 'Machu Picchu', 'Peru', 500000.00), (2, 'Great Wall', 'China', 700000.00), (3, 'Petra', 'Jordan', 600000.00);	database_business_operations	database_business_operations.database_query_analysis	heuristic_unreviewed	Mapped from source sql_task_type='analytics and reporting'; no individual SQL execution or semantic review.	en	none	SELECT Name, Location FROM (SELECT Name, Location, ROW_NUMBER() OVER (ORDER BY TotalDonations DESC) as rn FROM HeritageSites) t WHERE rn <= 3;	{"domain":"cultural preservation","domain_description":"In-depth data on heritage sites, traditional arts, language preservation, and community engagement in cultural preservation.","id":762,"sql_complexity":"window functions","sql_complexity_description":"window functions (e.g., ROW_NUMBER, LEAD, LAG, RANk, NTILE, PER...	dc12336decfa75a4d789ecda6789b4afbc64107f621a9e1c38da7b19cc8eb9ed	gretel_sql:test:762	window functions	intermediate	Estimated from source sql_complexity='window functions'; query feature labels are only proxies for difficulty, not measured solver performance. No empirical solver calibration was performed.	heuristic	needs_difficulty_calibration	not_measured	analysis_only_unvalidated	primary	task_annotation	761	[ "difficulty_not_calibrated", "not_individually_reviewed", "reference_not_independently_verified", "reference_or_constraint_only_no_rubric", "response_diversity_not_validated", "source_evaluation_split", "sql_dialect_and_result_equivalence_need_review", "sql_not_executed", "synthetic_reference" ]	false	true	false	task:bc95ae27ab56e584f96e47115c558f8119bd5a8b666d2592292d0451cbaa4462	1	1	[ "gretel_sql" ]	held_out	curated_candidate	[]	core_candidate	[ "source_sql_complexity:window functions", "sql_structure:selects=2;joins=0;statements=1;advanced=OVER", "combined_relational_operations_require_calibration" ]	structural_heuristic	Qwen3.6-27B	heuristic_not_model_measured
gretel_sql:5a8aef0121a287c4b608a17e724ff816cd28d47ed00cd80665509e2f10a03707	task:9aac2c920b3e6b48701b3a0fb8d316135bb4e5b060787e428ce3ac3eaf5a54e9	2c473044d22c2b9d5ab5eccfbf0b27573300dc08a81f65e2fba98229985c57f6	gretel_sql	gretelai/synthetic_text_to_sql	740ab236e64503fba51be1101df7a1be83bf455d	synthetic_text_to_sql_test.snappy.parquet	784	test	https://huggingface.co/datasets/gretelai/synthetic_text_to_sql/resolve/740ab236e64503fba51be1101df7a1be83bf455d/synthetic_text_to_sql_test.snappy.parquet	apache-2.0	Gretel Synthetic Text-to-SQL dataset creators; source card and author citation preserved in the download manifest.	proposed	List the top 2 countries with the highest water consumption in the current month.	CREATE TABLE water_consumption (country VARCHAR(255), consumption FLOAT, date DATE); INSERT INTO water_consumption (country, consumption, date) VALUES ('Brazil', 20000, '2022-05-01'); INSERT INTO water_consumption (country, consumption, date) VALUES ('Egypt', 25000, '2022-05-01');	database_business_operations	database_business_operations.database_query_analysis	heuristic_unreviewed	Mapped from source sql_task_type='analytics and reporting'; no individual SQL execution or semantic review.	en	none	SELECT country, consumption FROM (SELECT country, consumption, ROW_NUMBER() OVER (ORDER BY consumption DESC) as rank FROM water_consumption WHERE date >= '2022-05-01' GROUP BY country, consumption) subquery WHERE rank <= 2;	{"domain":"water resources","domain_description":"Water usage metrics, drought impact assessments, wastewater treatment data, and water conservation initiatives.","id":784,"sql_complexity":"window functions","sql_complexity_description":"window functions (e.g., ROW_NUMBER, LEAD, LAG, RANk, NTILE, PERCENT_RANK, etc.) wi...	dd3f1a80e263ba18813c8c88581f6d56f165958216c1b8f5667969b39b4eadc3	gretel_sql:test:784	window functions	intermediate	Estimated from source sql_complexity='window functions'; query feature labels are only proxies for difficulty, not measured solver performance. No empirical solver calibration was performed.	heuristic	needs_difficulty_calibration	not_measured	analysis_only_unvalidated	primary	task_annotation	783	[ "difficulty_not_calibrated", "not_individually_reviewed", "reference_not_independently_verified", "reference_or_constraint_only_no_rubric", "response_diversity_not_validated", "source_evaluation_split", "sql_dialect_and_result_equivalence_need_review", "sql_not_executed", "synthetic_reference" ]	false	true	false	task:9aac2c920b3e6b48701b3a0fb8d316135bb4e5b060787e428ce3ac3eaf5a54e9	1	1	[ "gretel_sql" ]	held_out	curated_candidate	[]	core_candidate	[ "source_sql_complexity:window functions", "sql_structure:selects=2;joins=0;statements=1;advanced=OVER", "combined_relational_operations_require_calibration" ]	structural_heuristic	Qwen3.6-27B	heuristic_not_model_measured
gretel_sql:cf559e5c2bae52d4272c5eb38a7e2167eff926c33014044282561ff7c31bab38	task:a38fc41269ee25b90f01c3a1d518e1f7d2ac103125ed9f6c2662086d957ce00c	31b14284a2446b7f5887d374afa6710847564bf3eb09d99d233307779b8f9d5e	gretel_sql	gretelai/synthetic_text_to_sql	740ab236e64503fba51be1101df7a1be83bf455d	synthetic_text_to_sql_test.snappy.parquet	812	test	https://huggingface.co/datasets/gretelai/synthetic_text_to_sql/resolve/740ab236e64503fba51be1101df7a1be83bf455d/synthetic_text_to_sql_test.snappy.parquet	apache-2.0	Gretel Synthetic Text-to-SQL dataset creators; source card and author citation preserved in the download manifest.	proposed	What is the average salary of 'engineer' workers in each factory?	CREATE TABLE factories (factory_id INT, factory_name VARCHAR(20)); INSERT INTO factories VALUES (1, 'Factory X'), (2, 'Factory Y'), (3, 'Factory Z'); CREATE TABLE roles (role_id INT, role_name VARCHAR(20)); INSERT INTO roles VALUES (1, 'engineer'), (2, 'manager'), (3, 'assistant'); CREATE TABLE workers (worker_id INT, ...	database_business_operations	database_business_operations.database_query_analysis	heuristic_unreviewed	Mapped from source sql_task_type='analytics and reporting'; no individual SQL execution or semantic review.	en	none	SELECT f.factory_name, AVG(salary) FROM workers w INNER JOIN factories f ON w.factory_id = f.factory_id INNER JOIN roles r ON w.role_id = r.role_id WHERE r.role_name = 'engineer' GROUP BY f.factory_name;	{"domain":"manufacturing","domain_description":"Detailed records on ethical manufacturing, circular economy, workforce development, and industry 4.0.","id":812,"sql_complexity":"multiple_joins","sql_complexity_description":"two or more joins (specify inner, outer, cross)","sql_explanation":"We join the 'workers', 'fact...	32729a02f28933aa612841bfc8d1614eebf512675b185150d8c3e7dd81e6c641	gretel_sql:test:812	multiple_joins	intermediate	Estimated from source sql_complexity='multiple_joins'; query feature labels are only proxies for difficulty, not measured solver performance. No empirical solver calibration was performed.	heuristic	needs_difficulty_calibration	not_measured	analysis_only_unvalidated	primary	task_annotation	811	[ "difficulty_not_calibrated", "not_individually_reviewed", "reference_not_independently_verified", "reference_or_constraint_only_no_rubric", "response_diversity_not_validated", "source_evaluation_split", "sql_dialect_and_result_equivalence_need_review", "sql_not_executed", "synthetic_reference" ]	false	true	false	task:a38fc41269ee25b90f01c3a1d518e1f7d2ac103125ed9f6c2662086d957ce00c	1	1	[ "gretel_sql" ]	held_out	curated_candidate	[]	core_candidate	[ "source_sql_complexity:multiple_joins", "sql_structure:selects=1;joins=2;statements=1;advanced=none", "combined_relational_operations_require_calibration" ]	structural_heuristic	Qwen3.6-27B	heuristic_not_model_measured
gretel_sql:465e31d4932a5b02f83c2cf7b9f54f2a2196e23eb394402f6de197940f6ec0ee	task:e5574d8b9ccbfa0e6932d295d20b7579909c7d6875a9c85bb97894b516ff0538	b032f9a26c6c0cedb5604b01ab4d101f953894fb7ce4811deec6cda5ec1ab148	gretel_sql	gretelai/synthetic_text_to_sql	740ab236e64503fba51be1101df7a1be83bf455d	synthetic_text_to_sql_test.snappy.parquet	827	test	https://huggingface.co/datasets/gretelai/synthetic_text_to_sql/resolve/740ab236e64503fba51be1101df7a1be83bf455d/synthetic_text_to_sql_test.snappy.parquet	apache-2.0	Gretel Synthetic Text-to-SQL dataset creators; source card and author citation preserved in the download manifest.	proposed	What is the average price of free-range eggs per store?	CREATE TABLE Stores (store_id INT, store_name VARCHAR(255)); CREATE TABLE Products (product_id INT, product_name VARCHAR(255), is_free_range BOOLEAN, price INT); CREATE TABLE Inventory (store_id INT, product_id INT, quantity INT);	database_business_operations	database_business_operations.database_query_analysis	heuristic_unreviewed	Mapped from source sql_task_type='analytics and reporting'; no individual SQL execution or semantic review.	en	none	SELECT s.store_name, AVG(p.price) as avg_price FROM Inventory i JOIN Stores s ON i.store_id = s.store_id JOIN Products p ON i.product_id = p.product_id WHERE p.is_free_range = TRUE AND p.product_category = 'egg' GROUP BY s.store_name;	{"domain":"food industry","domain_description":"Food safety records, nutrition data, supply chain transparency, and sustainable food trends.","id":827,"sql_complexity":"multiple_joins","sql_complexity_description":"two or more joins (specify inner, outer, cross)","sql_explanation":"This query joins the Stores, Products...	9ab947c4d6bc9c13812d17392c9bff770e34780b806291d6e034a1a082700f0d	gretel_sql:test:827	multiple_joins	intermediate	Estimated from source sql_complexity='multiple_joins'; query feature labels are only proxies for difficulty, not measured solver performance. No empirical solver calibration was performed.	heuristic	needs_difficulty_calibration	not_measured	analysis_only_unvalidated	primary	task_annotation	826	[ "difficulty_not_calibrated", "not_individually_reviewed", "reference_not_independently_verified", "reference_or_constraint_only_no_rubric", "response_diversity_not_validated", "source_evaluation_split", "sql_dialect_and_result_equivalence_need_review", "sql_not_executed", "synthetic_reference" ]	false	true	false	task:e5574d8b9ccbfa0e6932d295d20b7579909c7d6875a9c85bb97894b516ff0538	1	1	[ "gretel_sql" ]	held_out	curated_candidate	[]	core_candidate	[ "source_sql_complexity:multiple_joins", "sql_structure:selects=1;joins=2;statements=1;advanced=none", "combined_relational_operations_require_calibration" ]	structural_heuristic	Qwen3.6-27B	heuristic_not_model_measured
gretel_sql:5c30989fbb666c1be52f42e17ae1853c362200c2c002627ee8914eca4859c8fa	task:14368e576c6c19a25ef9a785dfa5cb7a3c23b2244503755d1b4eed6cbad9a939	ea117b6e90849d900c0ffc580e759d7119323088f4f77bd78e4d83848387abf1	gretel_sql	gretelai/synthetic_text_to_sql	740ab236e64503fba51be1101df7a1be83bf455d	synthetic_text_to_sql_test.snappy.parquet	895	test	https://huggingface.co/datasets/gretelai/synthetic_text_to_sql/resolve/740ab236e64503fba51be1101df7a1be83bf455d/synthetic_text_to_sql_test.snappy.parquet	apache-2.0	Gretel Synthetic Text-to-SQL dataset creators; source card and author citation preserved in the download manifest.	proposed	How many volunteers with 'Medical' skills were assigned before a volunteer with 'Engineering' skills?	CREATE TABLE volunteers_ext (id INT, name VARCHAR(50), age INT, gender VARCHAR(10), skill VARCHAR(50), assignment_date DATE, end_date DATE); INSERT INTO volunteers_ext (id, name, age, gender, skill, assignment_date, end_date) VALUES (1, 'David', 25, 'Male', 'Medical', '2022-06-01', '2022-09-30'), (2, 'Emma', 30, 'Femal...	database_business_operations	database_business_operations.database_query_analysis	heuristic_unreviewed	Mapped from source sql_task_type='analytics and reporting'; no individual SQL execution or semantic review.	en	none	SELECT COUNT(*) FROM (SELECT skill, assignment_date, LAG(skill) OVER (ORDER BY assignment_date) AS prev_skill FROM volunteers_ext WHERE skill = 'Medical') t WHERE prev_skill = 'Engineering';	{"domain":"humanitarian aid","domain_description":"Extensive data on disaster response, refugee support, community development, and advocacy in humanitarian aid.","id":895,"sql_complexity":"window functions","sql_complexity_description":"window functions (e.g., ROW_NUMBER, LEAD, LAG, RANk, NTILE, PERCENT_RANK, etc.) wi...	b4316fb2f1c4cbf3f49ec9be60740aec4eec75ef6cdb72203c6c1c3cbe66c59a	gretel_sql:test:895	window functions	intermediate	Estimated from source sql_complexity='window functions'; query feature labels are only proxies for difficulty, not measured solver performance. No empirical solver calibration was performed.	heuristic	needs_difficulty_calibration	not_measured	analysis_only_unvalidated	primary	task_annotation	894	[ "difficulty_not_calibrated", "not_individually_reviewed", "reference_not_independently_verified", "reference_or_constraint_only_no_rubric", "response_diversity_not_validated", "source_evaluation_split", "sql_dialect_and_result_equivalence_need_review", "sql_not_executed", "synthetic_reference" ]	false	true	false	task:14368e576c6c19a25ef9a785dfa5cb7a3c23b2244503755d1b4eed6cbad9a939	1	1	[ "gretel_sql" ]	held_out	curated_candidate	[]	core_candidate	[ "source_sql_complexity:window functions", "sql_structure:selects=2;joins=0;statements=1;advanced=OVER", "combined_relational_operations_require_calibration" ]	structural_heuristic	Qwen3.6-27B	heuristic_not_model_measured
gretel_sql:e59724c100af36653e60a79bf51e03365c936fa2e34be9f3358770f744c647a0	task:192a1ff1c67b2582f4c2bfbfe2b8a6d4e3361b1ee963fabc1f0bb516ff43483f	9b1643b005f623c0e0cfb0687a7fb79c34fc2fc9ca0b79c7cea94cfe46b32d50	gretel_sql	gretelai/synthetic_text_to_sql	740ab236e64503fba51be1101df7a1be83bf455d	synthetic_text_to_sql_test.snappy.parquet	926	test	https://huggingface.co/datasets/gretelai/synthetic_text_to_sql/resolve/740ab236e64503fba51be1101df7a1be83bf455d/synthetic_text_to_sql_test.snappy.parquet	apache-2.0	Gretel Synthetic Text-to-SQL dataset creators; source card and author citation preserved in the download manifest.	proposed	What is the average age of players who play VR games and their total spending on games?	CREATE TABLE players (id INT, name VARCHAR(50), age INT, country VARCHAR(50)); INSERT INTO players (id, name, age, country) VALUES (1, 'John Doe', 25, 'USA'), (2, 'Jane Smith', 30, 'Canada'); CREATE TABLE games (id INT, name VARCHAR(50), type VARCHAR(50), price DECIMAL(5,2)); INSERT INTO games (id, name, type, price) V...	database_business_operations	database_business_operations.database_query_analysis	heuristic_unreviewed	Mapped from source sql_task_type='analytics and reporting'; no individual SQL execution or semantic review.	en	none	SELECT AVG(players.age), SUM(games.price) FROM players INNER JOIN player_games ON players.id = player_games.player_id INNER JOIN games ON player_games.game_id = games.id WHERE games.type = 'VR';	{"domain":"gaming technology","domain_description":"Player demographics, game design data, esports event information, and virtual reality technology adoption.","id":926,"sql_complexity":"multiple_joins","sql_complexity_description":"two or more joins (specify inner, outer, cross)","sql_explanation":"First, we join the ...	ae7f531a5f3bcddad69b00e3a3047e65715ec65e2d71f77ed893bcc4a73dcf0b	gretel_sql:test:926	multiple_joins	intermediate	Estimated from source sql_complexity='multiple_joins'; query feature labels are only proxies for difficulty, not measured solver performance. No empirical solver calibration was performed.	heuristic	needs_difficulty_calibration	not_measured	analysis_only_unvalidated	primary	task_annotation	925	[ "difficulty_not_calibrated", "not_individually_reviewed", "reference_not_independently_verified", "reference_or_constraint_only_no_rubric", "response_diversity_not_validated", "source_evaluation_split", "sql_dialect_and_result_equivalence_need_review", "sql_not_executed", "synthetic_reference" ]	false	true	false	task:192a1ff1c67b2582f4c2bfbfe2b8a6d4e3361b1ee963fabc1f0bb516ff43483f	1	1	[ "gretel_sql" ]	held_out	curated_candidate	[]	core_candidate	[ "source_sql_complexity:multiple_joins", "sql_structure:selects=1;joins=2;statements=1;advanced=none", "combined_relational_operations_require_calibration" ]	structural_heuristic	Qwen3.6-27B	heuristic_not_model_measured
gretel_sql:66f2d9803d8a04db0de3ff55a25b95ab916be000a2c54667d4c4c704de5a640b	task:f35523c79d36444f3d64cd5403cdfa72c5c0c4530c2d32d186ec143f3f70dae0	4eb7871388f934fc562f451545b5fd3c1213a1c386738efd15926f514e332632	gretel_sql	gretelai/synthetic_text_to_sql	740ab236e64503fba51be1101df7a1be83bf455d	synthetic_text_to_sql_test.snappy.parquet	959	test	https://huggingface.co/datasets/gretelai/synthetic_text_to_sql/resolve/740ab236e64503fba51be1101df7a1be83bf455d/synthetic_text_to_sql_test.snappy.parquet	apache-2.0	Gretel Synthetic Text-to-SQL dataset creators; source card and author citation preserved in the download manifest.	proposed	What is the total salary paid to construction workers who worked on sustainable building projects in Washington?	CREATE TABLE ConstructionLaborStatistics (id INT, name VARCHAR(50), job VARCHAR(50), salary INT); INSERT INTO ConstructionLaborStatistics VALUES (1, 'Charles Doe', 'Carpenter', 50000); INSERT INTO ConstructionLaborStatistics VALUES (2, 'Diana Smith', 'Electrician', 60000); CREATE TABLE BuildingTypes (id INT, building_t...	database_business_operations	database_business_operations.database_query_analysis	heuristic_unreviewed	Mapped from source sql_task_type='analytics and reporting'; no individual SQL execution or semantic review.	en	none	SELECT SUM(cls.salary) FROM ConstructionLaborStatistics cls JOIN WorkerBuildings wb ON cls.id = wb.worker_id JOIN BuildingTypes bt ON wb.building_id = bt.id WHERE bt.building_type = 'Sustainable' AND state = 'Washington';	{"domain":"construction","domain_description":"Building permit data, construction labor statistics, project timeline information, and sustainable building practices.","id":959,"sql_complexity":"multiple_joins","sql_complexity_description":"two or more joins (specify inner, outer, cross)","sql_explanation":"We find the ...	46e6728bd1a85e027a69d5fbe73a362e9a4489b37fee9efb6110ea8af635a8d6	gretel_sql:test:959	multiple_joins	intermediate	Estimated from source sql_complexity='multiple_joins'; query feature labels are only proxies for difficulty, not measured solver performance. No empirical solver calibration was performed.	heuristic	needs_difficulty_calibration	not_measured	analysis_only_unvalidated	primary	task_annotation	958	[ "difficulty_not_calibrated", "not_individually_reviewed", "reference_not_independently_verified", "reference_or_constraint_only_no_rubric", "response_diversity_not_validated", "source_evaluation_split", "sql_dialect_and_result_equivalence_need_review", "sql_not_executed", "synthetic_reference" ]	false	true	false	task:f35523c79d36444f3d64cd5403cdfa72c5c0c4530c2d32d186ec143f3f70dae0	1	1	[ "gretel_sql" ]	held_out	curated_candidate	[]	core_candidate	[ "source_sql_complexity:multiple_joins", "sql_structure:selects=1;joins=2;statements=1;advanced=none", "combined_relational_operations_require_calibration" ]	structural_heuristic	Qwen3.6-27B	heuristic_not_model_measured
gretel_sql:c2c8897bca9564cfd28a876124d3931163624130d248f21cfb497c72edefc231	task:e7b4b5f2110113e5ecdac39ac67ff02aacbca58c5363f23bd1725f7dd2776d6b	64847034182786e5eabd96e3518f7a07e67c7aaaf5ecf761d788eb10781dd8e8	gretel_sql	gretelai/synthetic_text_to_sql	740ab236e64503fba51be1101df7a1be83bf455d	synthetic_text_to_sql_test.snappy.parquet	978	test	https://huggingface.co/datasets/gretelai/synthetic_text_to_sql/resolve/740ab236e64503fba51be1101df7a1be83bf455d/synthetic_text_to_sql_test.snappy.parquet	apache-2.0	Gretel Synthetic Text-to-SQL dataset creators; source card and author citation preserved in the download manifest.	proposed	What is the production rate for the well with the highest production rate?	CREATE TABLE wells (well_id INT, well_type VARCHAR(10), location VARCHAR(20), production_rate FLOAT); INSERT INTO wells (well_id, well_type, location, production_rate) VALUES (1, 'offshore', 'Gulf of Mexico', 1000), (2, 'onshore', 'Texas', 800), (3, 'offshore', 'North Sea', 1200);	database_business_operations	database_business_operations.database_query_analysis	heuristic_unreviewed	Mapped from source sql_task_type='analytics and reporting'; no individual SQL execution or semantic review.	en	none	SELECT production_rate FROM (SELECT well_id, well_type, location, production_rate, ROW_NUMBER() OVER (ORDER BY production_rate DESC) rn FROM wells) t WHERE rn = 1;	{"domain":"oil and gas","domain_description":"Exploration data, production figures, infrastructure development, and market trends.","id":978,"sql_complexity":"window functions","sql_complexity_description":"window functions (e.g., ROW_NUMBER, LEAD, LAG, RANk, NTILE, PERCENT_RANK, etc.) with partitioning and ordering","...	2a5cfa63434192e8899a6d24c32b5e41eb96e71716ba44dab4f8f5c7946f172e	gretel_sql:test:978	window functions	intermediate	Estimated from source sql_complexity='window functions'; query feature labels are only proxies for difficulty, not measured solver performance. No empirical solver calibration was performed.	heuristic	needs_difficulty_calibration	not_measured	analysis_only_unvalidated	primary	task_annotation	977	[ "difficulty_not_calibrated", "not_individually_reviewed", "reference_not_independently_verified", "reference_or_constraint_only_no_rubric", "response_diversity_not_validated", "source_evaluation_split", "sql_dialect_and_result_equivalence_need_review", "sql_not_executed", "synthetic_reference" ]	false	true	false	task:e7b4b5f2110113e5ecdac39ac67ff02aacbca58c5363f23bd1725f7dd2776d6b	1	1	[ "gretel_sql" ]	held_out	curated_candidate	[]	core_candidate	[ "source_sql_complexity:window functions", "sql_structure:selects=2;joins=0;statements=1;advanced=OVER", "combined_relational_operations_require_calibration" ]	structural_heuristic	Qwen3.6-27B	heuristic_not_model_measured
gretel_sql:57dd8407cbca173b558f00ef01f253e505f13cdf433ee055ae0e5513271eba2b	task:54457ae6a4f1c948787157d5269384e546075af95d8e601a04c1b46621dcab9b	e76a2d0a6305cc0f378b6f90f7fee120d8128a87c44970567c59aecb3ccbf083	gretel_sql	gretelai/synthetic_text_to_sql	740ab236e64503fba51be1101df7a1be83bf455d	synthetic_text_to_sql_test.snappy.parquet	1009	test	https://huggingface.co/datasets/gretelai/synthetic_text_to_sql/resolve/740ab236e64503fba51be1101df7a1be83bf455d/synthetic_text_to_sql_test.snappy.parquet	apache-2.0	Gretel Synthetic Text-to-SQL dataset creators; source card and author citation preserved in the download manifest.	proposed	Find the number of IoT sensors installed in each farm that use Sprinkler irrigation.	CREATE TABLE IoT_Sensors (id INT, sensor_type VARCHAR(50), Farm_id INT); INSERT INTO IoT_Sensors (id, sensor_type, Farm_id) VALUES (1, 'Soil Moisture', 1), (2, 'Temperature', 1), (3, 'Humidity', 2); CREATE TABLE Irrigation (id INT, Farm_id INT, irrigation_type VARCHAR(50), duration INT); INSERT INTO Irrigation (id, Far...	database_business_operations	database_business_operations.database_query_analysis	heuristic_unreviewed	Mapped from source sql_task_type='analytics and reporting'; no individual SQL execution or semantic review.	en	none	SELECT f.id, COUNT(s.id) FROM Farmers f JOIN Irrigation i ON f.id = i.Farm_id JOIN IoT_Sensors s ON f.id = s.Farm_id WHERE i.irrigation_type = 'Sprinkler' GROUP BY f.id;	{"domain":"precision agriculture","domain_description":"Precision farming data, satellite imagery analysis, IoT sensor metrics, and agricultural automation trends.","id":1009,"sql_complexity":"multiple_joins","sql_complexity_description":"two or more joins (specify inner, outer, cross)","sql_explanation":"This query jo...	18e0c6a04e73d57937e51b8f42dce495bada1fcce20ed6453d850d566f9d2c13	gretel_sql:test:1009	multiple_joins	intermediate	Estimated from source sql_complexity='multiple_joins'; query feature labels are only proxies for difficulty, not measured solver performance. No empirical solver calibration was performed.	heuristic	needs_difficulty_calibration	not_measured	analysis_only_unvalidated	primary	task_annotation	1,008	[ "difficulty_not_calibrated", "not_individually_reviewed", "reference_not_independently_verified", "reference_or_constraint_only_no_rubric", "response_diversity_not_validated", "source_evaluation_split", "sql_dialect_and_result_equivalence_need_review", "sql_not_executed", "synthetic_reference" ]	false	true	false	task:54457ae6a4f1c948787157d5269384e546075af95d8e601a04c1b46621dcab9b	1	1	[ "gretel_sql" ]	held_out	curated_candidate	[]	core_candidate	[ "source_sql_complexity:multiple_joins", "sql_structure:selects=1;joins=2;statements=1;advanced=none", "combined_relational_operations_require_calibration" ]	structural_heuristic	Qwen3.6-27B	heuristic_not_model_measured
gretel_sql:a3a6f3a5ff71f557ad8867b5153fb65fd983b8aceba08e05ee2c1141f7b135b5	task:f979f8ef3f1bfee4bc4e9af1863d72ea3c95c88873dc82faf56a9a6cc0bc2e44	b5f416338ad2926d1bdffdfb727cf11827a3ed3e1588d73aed0ea6e2ab062e19	gretel_sql	gretelai/synthetic_text_to_sql	740ab236e64503fba51be1101df7a1be83bf455d	synthetic_text_to_sql_test.snappy.parquet	1033	test	https://huggingface.co/datasets/gretelai/synthetic_text_to_sql/resolve/740ab236e64503fba51be1101df7a1be83bf455d/synthetic_text_to_sql_test.snappy.parquet	apache-2.0	Gretel Synthetic Text-to-SQL dataset creators; source card and author citation preserved in the download manifest.	proposed	What is the average dissolved oxygen level for each species in our fish farms?	CREATE TABLE fish_farms (id INT, name TEXT, location TEXT, water_type TEXT); INSERT INTO fish_farms (id, name, location, water_type) VALUES (1, 'Farm A', 'Seattle', 'Saltwater'); INSERT INTO fish_farms (id, name, location, water_type) VALUES (2, 'Farm B', 'Portland', 'Freshwater'); CREATE TABLE fish_species (id INT, na...	database_business_operations	database_business_operations.database_query_analysis	heuristic_unreviewed	Mapped from source sql_task_type='analytics and reporting'; no individual SQL execution or semantic review.	en	none	SELECT fs.name AS species_name, AVG(av.dissolved_oxygen) AS avg_dissolved_oxygen FROM fish_inventory fi JOIN fish_farms ff ON fi.fish_farm_id = ff.id JOIN fish_species fs ON fi.fish_species_id = fs.id JOIN (SELECT fish_species_id, AVG(dissolved_oxygen) AS dissolved_oxygen FROM water_quality GROUP BY fish_species_id) av...	{"domain":"aquaculture","domain_description":"Aquatic farming data, fish stock management, ocean health metrics, and sustainable seafood trends.","id":1033,"sql_complexity":"multiple_joins","sql_complexity_description":"two or more joins (specify inner, outer, cross)","sql_explanation":"This query joins the fish_farms,...	887ab0c78c632b519f1071f7119b01202ba110c11be6cb56df2fa497de06fbcf	gretel_sql:test:1033	multiple_joins	intermediate	Estimated from source sql_complexity='multiple_joins'; query feature labels are only proxies for difficulty, not measured solver performance. No empirical solver calibration was performed.	heuristic	needs_difficulty_calibration	not_measured	analysis_only_unvalidated	primary	task_annotation	1,032	[ "difficulty_not_calibrated", "not_individually_reviewed", "reference_not_independently_verified", "reference_or_constraint_only_no_rubric", "response_diversity_not_validated", "source_evaluation_split", "sql_dialect_and_result_equivalence_need_review", "sql_not_executed", "synthetic_reference" ]	false	true	false	task:f979f8ef3f1bfee4bc4e9af1863d72ea3c95c88873dc82faf56a9a6cc0bc2e44	1	1	[ "gretel_sql" ]	held_out	curated_candidate	[]	core_candidate	[ "source_sql_complexity:multiple_joins", "sql_structure:selects=2;joins=3;statements=1;advanced=none", "combined_relational_operations_require_calibration" ]	structural_heuristic	Qwen3.6-27B	heuristic_not_model_measured
gretel_sql:f9b9d0cbd6b8f356f1fdb0252b797cfb5cdefc97d635773b95d85c2688076955	task:418483412367aad348001ae1e866b409432f34c35a911db3427878c1b44ea09e	647fc6fd9339004874dcc92dff31fca841adc36ed2dd74f99df831b6177ea05d	gretel_sql	gretelai/synthetic_text_to_sql	740ab236e64503fba51be1101df7a1be83bf455d	synthetic_text_to_sql_test.snappy.parquet	1126	test	https://huggingface.co/datasets/gretelai/synthetic_text_to_sql/resolve/740ab236e64503fba51be1101df7a1be83bf455d/synthetic_text_to_sql_test.snappy.parquet	apache-2.0	Gretel Synthetic Text-to-SQL dataset creators; source card and author citation preserved in the download manifest.	proposed	How many innovations have been made in the African region since 2016?	CREATE TABLE region (id INT, region VARCHAR(50)); INSERT INTO region (id, region) VALUES (1, 'North America'); INSERT INTO region (id, region) VALUES (2, 'Europe'); INSERT INTO region (id, region) VALUES (3, 'Africa'); CREATE TABLE innovation_region (id INT, innovation_id INT, region_id INT); INSERT INTO innovation_reg...	database_business_operations	database_business_operations.database_query_analysis	heuristic_unreviewed	Mapped from source sql_task_type='analytics and reporting'; no individual SQL execution or semantic review.	en	none	SELECT COUNT(*) FROM innovation i INNER JOIN innovation_region ir ON i.id = ir.innovation_id INNER JOIN region r ON ir.region_id = r.id WHERE r.region = 'Africa' AND i.year >= 2016;	{"domain":"startups","domain_description":"Company founding data, funding records, diversity metrics, and innovation trends.","id":1126,"sql_complexity":"multiple_joins","sql_complexity_description":"two or more joins (specify inner, outer, cross)","sql_explanation":"This query joins the innovation, innovation_region, ...	e583dfd8a9b9149ee7b6055b732bfb8cb6e5a343d36d60b42ce759b947dc3ef2	gretel_sql:test:1126	multiple_joins	intermediate	Estimated from source sql_complexity='multiple_joins'; query feature labels are only proxies for difficulty, not measured solver performance. No empirical solver calibration was performed.	heuristic	needs_difficulty_calibration	not_measured	analysis_only_unvalidated	primary	task_annotation	1,125	[ "difficulty_not_calibrated", "not_individually_reviewed", "reference_not_independently_verified", "reference_or_constraint_only_no_rubric", "response_diversity_not_validated", "source_evaluation_split", "sql_dialect_and_result_equivalence_need_review", "sql_not_executed", "synthetic_reference" ]	false	true	false	task:418483412367aad348001ae1e866b409432f34c35a911db3427878c1b44ea09e	1	1	[ "gretel_sql" ]	held_out	curated_candidate	[]	core_candidate	[ "source_sql_complexity:multiple_joins", "sql_structure:selects=1;joins=2;statements=1;advanced=none", "combined_relational_operations_require_calibration" ]	structural_heuristic	Qwen3.6-27B	heuristic_not_model_measured
gretel_sql:c69e12dbc5c450a9368a5b805663e84ea5abaa96216b9e58c14daa243739ab5d	task:dc854f72eecca7b38213372fe5935fcae76b70f22149574bae5d2ca90ad6efd1	cb5f90d914b6709023a07b1e55e3f7b3166b063f4fa034702bfeb357a361dc8c	gretel_sql	gretelai/synthetic_text_to_sql	740ab236e64503fba51be1101df7a1be83bf455d	synthetic_text_to_sql_test.snappy.parquet	1178	test	https://huggingface.co/datasets/gretelai/synthetic_text_to_sql/resolve/740ab236e64503fba51be1101df7a1be83bf455d/synthetic_text_to_sql_test.snappy.parquet	apache-2.0	Gretel Synthetic Text-to-SQL dataset creators; source card and author citation preserved in the download manifest.	proposed	List the booking dates and hotel names for all OTA bookings where the hotel has implemented at least one AI-powered solution.	CREATE TABLE otas (ota_id INT, booking_date DATE, hotel_id INT); CREATE TABLE hotels (hotel_id INT, hotel_name TEXT, region TEXT); CREATE TABLE ai_solutions (solution_id INT, hotel_id INT, implemented_date DATE); INSERT INTO hotels (hotel_id, hotel_name, region) VALUES (1, 'Beach Retreat', 'Americas'); INSERT INTO ai_s...	database_business_operations	database_business_operations.database_query_analysis	heuristic_unreviewed	Mapped from source sql_task_type='analytics and reporting'; no individual SQL execution or semantic review.	en	none	SELECT otas.booking_date, hotels.hotel_name FROM otas INNER JOIN hotels ON otas.hotel_id = hotels.hotel_id INNER JOIN ai_solutions ON hotels.hotel_id = ai_solutions.hotel_id GROUP BY otas.booking_date, hotels.hotel_name HAVING COUNT(DISTINCT ai_solutions.solution_id) >= 1;	{"domain":"hospitality technology","domain_description":"Hotel tech adoption metrics, online travel agency data, virtual tour engagement stats, and hospitality AI trends.","id":1178,"sql_complexity":"multiple_joins","sql_complexity_description":"two or more joins (specify inner, outer, cross)","sql_explanation":"This S...	cd33d91e040fb59fb6292d32bc51f67aa98f3b7ada08a6a7032bdeb02f805161	gretel_sql:test:1178	multiple_joins	intermediate	Estimated from source sql_complexity='multiple_joins'; query feature labels are only proxies for difficulty, not measured solver performance. No empirical solver calibration was performed.	heuristic	needs_difficulty_calibration	not_measured	analysis_only_unvalidated	primary	task_annotation	1,177	[ "difficulty_not_calibrated", "not_individually_reviewed", "reference_not_independently_verified", "reference_or_constraint_only_no_rubric", "response_diversity_not_validated", "source_evaluation_split", "sql_dialect_and_result_equivalence_need_review", "sql_not_executed", "synthetic_reference" ]	false	true	false	task:dc854f72eecca7b38213372fe5935fcae76b70f22149574bae5d2ca90ad6efd1	1	1	[ "gretel_sql" ]	held_out	curated_candidate	[]	core_candidate	[ "source_sql_complexity:multiple_joins", "sql_structure:selects=1;joins=2;statements=1;advanced=none", "combined_relational_operations_require_calibration" ]	structural_heuristic	Qwen3.6-27B	heuristic_not_model_measured
gretel_sql:8091ef2ad6bfd5f179cf52d7d9c5784b9752f35a18d8f92632ddfd3e7bcaadb2	task:b71d9e9e836386899ba8fab23e0e852d6ef781e10c294c226d79df57b32a5b48	e05fc109dd77313ecbe78a07430ab46389c6cd1b4f8e158edbcb3499b83784e7	gretel_sql	gretelai/synthetic_text_to_sql	740ab236e64503fba51be1101df7a1be83bf455d	synthetic_text_to_sql_test.snappy.parquet	1187	test	https://huggingface.co/datasets/gretelai/synthetic_text_to_sql/resolve/740ab236e64503fba51be1101df7a1be83bf455d/synthetic_text_to_sql_test.snappy.parquet	apache-2.0	Gretel Synthetic Text-to-SQL dataset creators; source card and author citation preserved in the download manifest.	proposed	Who are the top 3 authors with the highest number of articles published in The Guardian?	CREATE TABLE authors (id INT, name VARCHAR(100), publisher VARCHAR(50)); CREATE TABLE articles_authors (article_id INT, author_id INT); INSERT INTO authors (id, name, publisher) VALUES (1, 'Author1', 'The Guardian'), (2, 'Author2', 'The Guardian'), (3, 'Author3', 'The Guardian'); INSERT INTO articles_authors (article_i...	database_business_operations	database_business_operations.database_query_analysis	heuristic_unreviewed	Mapped from source sql_task_type='analytics and reporting'; no individual SQL execution or semantic review.	en	none	SELECT a.name, COUNT(aa.article_id) AS articles_count FROM authors a JOIN articles_authors aa ON a.id = aa.author_id JOIN articles ar ON aa.article_id = ar.id WHERE ar.publisher = 'The Guardian' GROUP BY a.name ORDER BY articles_count DESC LIMIT 3;	{"domain":"media","domain_description":"Media data on content diversity, media literacy, disinformation detection, and media representation.","id":1187,"sql_complexity":"multiple_joins","sql_complexity_description":"two or more joins (specify inner, outer, cross)","sql_explanation":"Find the top 3 authors with the high...	79566f440acc5280d1fe46233ad6ac4a5057421fb1ba103e44a2974000552a8d	gretel_sql:test:1187	multiple_joins	intermediate	Estimated from source sql_complexity='multiple_joins'; query feature labels are only proxies for difficulty, not measured solver performance. No empirical solver calibration was performed.	heuristic	needs_difficulty_calibration	not_measured	analysis_only_unvalidated	primary	task_annotation	1,186	[ "difficulty_not_calibrated", "not_individually_reviewed", "reference_not_independently_verified", "reference_or_constraint_only_no_rubric", "response_diversity_not_validated", "source_evaluation_split", "sql_dialect_and_result_equivalence_need_review", "sql_not_executed", "synthetic_reference" ]	false	true	false	task:b71d9e9e836386899ba8fab23e0e852d6ef781e10c294c226d79df57b32a5b48	1	1	[ "gretel_sql" ]	held_out	curated_candidate	[]	core_candidate	[ "source_sql_complexity:multiple_joins", "sql_structure:selects=1;joins=2;statements=1;advanced=none", "combined_relational_operations_require_calibration" ]	structural_heuristic	Qwen3.6-27B	heuristic_not_model_measured
gretel_sql:5ca9ec89e334c46262dc057ae5b3672513875e655f83e66402db9b6776e583a3	task:939eb10a689dc7ee5910af3e2631fc6fe324d187ca4b02e46aeb86cdde707339	fc1477d1e3b668dd55c4e4c79eac8415f9a19fce4d3f35ba6dbb9b08a67ab976	gretel_sql	gretelai/synthetic_text_to_sql	740ab236e64503fba51be1101df7a1be83bf455d	synthetic_text_to_sql_test.snappy.parquet	1286	test	https://huggingface.co/datasets/gretelai/synthetic_text_to_sql/resolve/740ab236e64503fba51be1101df7a1be83bf455d/synthetic_text_to_sql_test.snappy.parquet	apache-2.0	Gretel Synthetic Text-to-SQL dataset creators; source card and author citation preserved in the download manifest.	proposed	Identify the top 3 countries with the highest percentage of attendees	CREATE TABLE attendee_info (attendee_id INT, country VARCHAR(20)); INSERT INTO attendee_info (attendee_id, country) VALUES (1, 'USA'), (2, 'Canada'), (3, 'Mexico'), (4, 'USA'), (5, 'Brazil'), (6, 'USA');	database_business_operations	database_business_operations.database_query_analysis	heuristic_unreviewed	Mapped from source sql_task_type='analytics and reporting'; no individual SQL execution or semantic review.	en	none	SELECT country, (COUNT(attendee_id) OVER (PARTITION BY country) * 100.0 / (SELECT COUNT(attendee_id) FROM attendee_info)) AS percentage FROM attendee_info GROUP BY country ORDER BY percentage DESC LIMIT 3;	{"domain":"arts and culture","domain_description":"Audience demographics, event attendance, program impact, and funding sources.","id":1286,"sql_complexity":"window functions","sql_complexity_description":"window functions (e.g., ROW_NUMBER, LEAD, LAG, RANk, NTILE, PERCENT_RANK, etc.) with partitioning and ordering","s...	a3b27dcf054f2409ca3ef0b1252c1db16b374ee71f2b21f75eb45921ee5e4d6a	gretel_sql:test:1286	window functions	intermediate	Estimated from source sql_complexity='window functions'; query feature labels are only proxies for difficulty, not measured solver performance. No empirical solver calibration was performed.	heuristic	needs_difficulty_calibration	not_measured	analysis_only_unvalidated	primary	task_annotation	1,285	[ "difficulty_not_calibrated", "not_individually_reviewed", "reference_not_independently_verified", "reference_or_constraint_only_no_rubric", "response_diversity_not_validated", "source_evaluation_split", "sql_dialect_and_result_equivalence_need_review", "sql_not_executed", "synthetic_reference" ]	false	true	false	task:939eb10a689dc7ee5910af3e2631fc6fe324d187ca4b02e46aeb86cdde707339	1	1	[ "gretel_sql" ]	held_out	curated_candidate	[]	core_candidate	[ "source_sql_complexity:window functions", "sql_structure:selects=2;joins=0;statements=1;advanced=OVER", "combined_relational_operations_require_calibration" ]	structural_heuristic	Qwen3.6-27B	heuristic_not_model_measured
gretel_sql:8b6199f7be509bfdb02b69909153a02dbada923919ff778c44fc721fdfbc08e4	task:dd65ae24ccfeaf9deb50355ab01bd74336df6c7793e18da45ca0a997029cb989	029152ac5a4f73586513e210e1df3b68b88ad52b6db08a2fc1a4e85033d4646b	gretel_sql	gretelai/synthetic_text_to_sql	740ab236e64503fba51be1101df7a1be83bf455d	synthetic_text_to_sql_test.snappy.parquet	1301	test	https://huggingface.co/datasets/gretelai/synthetic_text_to_sql/resolve/740ab236e64503fba51be1101df7a1be83bf455d/synthetic_text_to_sql_test.snappy.parquet	apache-2.0	Gretel Synthetic Text-to-SQL dataset creators; source card and author citation preserved in the download manifest.	proposed	What is the total quantity of corn sold by farmers in 'Summerfield'?	CREATE TABLE farmers (id INT, name VARCHAR(50), location VARCHAR(50), crops VARCHAR(50)); CREATE TABLE crops (id INT, name VARCHAR(50), yield INT); CREATE TABLE sales (id INT, farmer_id INT, crop_name VARCHAR(50), quantity INT, price DECIMAL(5,2)); INSERT INTO farmers VALUES (1, 'Jane Doe', 'Summerfield', 'Corn'); INSE...	database_business_operations	database_business_operations.database_query_analysis	heuristic_unreviewed	Mapped from source sql_task_type='analytics and reporting'; no individual SQL execution or semantic review.	en	none	SELECT SUM(quantity) FROM sales INNER JOIN farmers ON sales.farmer_id = farmers.id INNER JOIN crops ON sales.crop_name = crops.name WHERE farmers.location = 'Summerfield' AND crops.name = 'Corn';	{"domain":"agriculture","domain_description":"Comprehensive data on agroecology, food justice, indigenous food systems, and urban agriculture.","id":1301,"sql_complexity":"multiple_joins","sql_complexity_description":"two or more joins (specify inner, outer, cross)","sql_explanation":"This query performs an inner join ...	86c05ef3c9666eccccdbc317eef1dc90fe593d5303f2cc9469be06b5f4ef38c1	gretel_sql:test:1301	multiple_joins	intermediate	Estimated from source sql_complexity='multiple_joins'; query feature labels are only proxies for difficulty, not measured solver performance. No empirical solver calibration was performed.	heuristic	needs_difficulty_calibration	not_measured	analysis_only_unvalidated	primary	task_annotation	1,300	[ "difficulty_not_calibrated", "not_individually_reviewed", "reference_not_independently_verified", "reference_or_constraint_only_no_rubric", "response_diversity_not_validated", "source_evaluation_split", "sql_dialect_and_result_equivalence_need_review", "sql_not_executed", "synthetic_reference" ]	false	true	false	task:dd65ae24ccfeaf9deb50355ab01bd74336df6c7793e18da45ca0a997029cb989	1	1	[ "gretel_sql" ]	held_out	curated_candidate	[]	core_candidate	[ "source_sql_complexity:multiple_joins", "sql_structure:selects=1;joins=2;statements=1;advanced=none", "combined_relational_operations_require_calibration" ]	structural_heuristic	Qwen3.6-27B	heuristic_not_model_measured
gretel_sql:87e28bf7631d6664e4f85539ca2bb5b6fd275199a3b872fbe5cdd24a2b15cb76	task:3339f470b71c58870339ba40b989859de73c49ab732efa8ee0a6a5c9a079ec6c	c136c6516cc57d182ac637418f3cb3abeef5347f144dec6ee9e50ca3c9b3dfd6	gretel_sql	gretelai/synthetic_text_to_sql	740ab236e64503fba51be1101df7a1be83bf455d	synthetic_text_to_sql_test.snappy.parquet	1303	test	https://huggingface.co/datasets/gretelai/synthetic_text_to_sql/resolve/740ab236e64503fba51be1101df7a1be83bf455d/synthetic_text_to_sql_test.snappy.parquet	apache-2.0	Gretel Synthetic Text-to-SQL dataset creators; source card and author citation preserved in the download manifest.	proposed	What is the average total value of transactions for the top 3 digital assets in the 'Binance Smart Chain' network?	CREATE TABLE binance_transactions (asset_name VARCHAR(20), network VARCHAR(20), transactions_value FLOAT); INSERT INTO binance_transactions (asset_name, network, transactions_value) VALUES ('BNB', 'Binance Smart Chain', 200000), ('ETH', 'Binance Smart Chain', 300000), ('CAKE', 'Binance Smart Chain', 400000);	database_business_operations	database_business_operations.database_query_analysis	heuristic_unreviewed	Mapped from source sql_task_type='analytics and reporting'; no individual SQL execution or semantic review.	en	none	SELECT asset_name, network, AVG(transactions_value) FROM binance_transactions WHERE network = 'Binance Smart Chain' AND asset_name IN (SELECT asset_name FROM (SELECT asset_name, ROW_NUMBER() OVER (ORDER BY transactions_value DESC) as rn FROM binance_transactions WHERE network = 'Binance Smart Chain') x WHERE rn <= 3) G...	{"domain":"blockchain","domain_description":"Comprehensive data on smart contracts, decentralized applications, digital assets, and regulatory frameworks in blockchain.","id":1303,"sql_complexity":"window functions","sql_complexity_description":"window functions (e.g., ROW_NUMBER, LEAD, LAG, RANk, NTILE, PERCENT_RANK, ...	1435f1b77e5a316065c917773514494a3324c255d991599c04e32fa67f4f0af8	gretel_sql:test:1303	window functions	intermediate	Estimated from source sql_complexity='window functions'; query feature labels are only proxies for difficulty, not measured solver performance. No empirical solver calibration was performed.	heuristic	needs_difficulty_calibration	not_measured	analysis_only_unvalidated	primary	task_annotation	1,302	[ "difficulty_not_calibrated", "not_individually_reviewed", "reference_not_independently_verified", "reference_or_constraint_only_no_rubric", "response_diversity_not_validated", "source_evaluation_split", "sql_dialect_and_result_equivalence_need_review", "sql_not_executed", "synthetic_reference" ]	false	true	false	task:3339f470b71c58870339ba40b989859de73c49ab732efa8ee0a6a5c9a079ec6c	1	1	[ "gretel_sql" ]	held_out	curated_candidate	[]	core_candidate	[ "source_sql_complexity:window functions", "sql_structure:selects=3;joins=0;statements=1;advanced=OVER", "combined_relational_operations_require_calibration" ]	structural_heuristic	Qwen3.6-27B	heuristic_not_model_measured
gretel_sql:4cded537725dc5ab510341218ef158624d6801f127e0e912c1b581a1a396cf46	task:f817c483a186b0c73ce39caa88db2575b17606e7e4845a17c6fd7ccc13acab63	50d7ad974cb3ec769e37adeafb1ff97c1fd59ed1a9d7dabfcd0a6cb6b7195069	gretel_sql	gretelai/synthetic_text_to_sql	740ab236e64503fba51be1101df7a1be83bf455d	synthetic_text_to_sql_test.snappy.parquet	1316	test	https://huggingface.co/datasets/gretelai/synthetic_text_to_sql/resolve/740ab236e64503fba51be1101df7a1be83bf455d/synthetic_text_to_sql_test.snappy.parquet	apache-2.0	Gretel Synthetic Text-to-SQL dataset creators; source card and author citation preserved in the download manifest.	proposed	List the ports and their average cargo weight for company "HarborLink" in Q2 2017 and Q3 2017.	CREATE TABLE company (id INT, name VARCHAR(255)); INSERT INTO company (id, name) VALUES (1, 'HarborLink'); CREATE TABLE port (id INT, name VARCHAR(255)); CREATE TABLE cargo (id INT, port_id INT, company_id INT, weight INT, quarter INT); INSERT INTO port (id, name) VALUES (1, 'PortA'), (2, 'PortB'), (3, 'PortC'); INSERT...	database_business_operations	database_business_operations.database_query_analysis	heuristic_unreviewed	Mapped from source sql_task_type='analytics and reporting'; no individual SQL execution or semantic review.	en	none	SELECT port.name, AVG(cargo.weight) FROM port INNER JOIN cargo ON port.id = cargo.port_id AND cargo.quarter IN (2, 3) INNER JOIN company ON cargo.company_id = company.id WHERE company.name = 'HarborLink' GROUP BY port.name;	{"domain":"ocean shipping","domain_description":"Detailed records on cargo handling, fleet management, port operations, and regulatory compliance in ocean shipping.","id":1316,"sql_complexity":"multiple_joins","sql_complexity_description":"two or more joins (specify inner, outer, cross)","sql_explanation":"This query c...	bbedad588f3a3452596d0c589c32a4ff72bc43fbeff7e4dfd24821c77866a9c0	gretel_sql:test:1316	multiple_joins	intermediate	Estimated from source sql_complexity='multiple_joins'; query feature labels are only proxies for difficulty, not measured solver performance. No empirical solver calibration was performed.	heuristic	needs_difficulty_calibration	not_measured	analysis_only_unvalidated	primary	task_annotation	1,315	[ "difficulty_not_calibrated", "not_individually_reviewed", "reference_not_independently_verified", "reference_or_constraint_only_no_rubric", "response_diversity_not_validated", "source_evaluation_split", "sql_dialect_and_result_equivalence_need_review", "sql_not_executed", "synthetic_reference" ]	false	true	false	task:f817c483a186b0c73ce39caa88db2575b17606e7e4845a17c6fd7ccc13acab63	1	1	[ "gretel_sql" ]	held_out	curated_candidate	[]	core_candidate	[ "source_sql_complexity:multiple_joins", "sql_structure:selects=1;joins=2;statements=1;advanced=none", "combined_relational_operations_require_calibration" ]	structural_heuristic	Qwen3.6-27B	heuristic_not_model_measured
gretel_sql:762a41d3d49035e6cdbda94976579a01299870ba0632587b6a5fa07369ca151c	task:cb9973e895539e119850c9d0aa0f408b8d18a46b81d99ae55ea5b246f03b7253	a0bc3b8291877c9331c13e939f06610a929cea7e3b75ee1b38dedc1715d32ad3	gretel_sql	gretelai/synthetic_text_to_sql	740ab236e64503fba51be1101df7a1be83bf455d	synthetic_text_to_sql_test.snappy.parquet	1321	test	https://huggingface.co/datasets/gretelai/synthetic_text_to_sql/resolve/740ab236e64503fba51be1101df7a1be83bf455d/synthetic_text_to_sql_test.snappy.parquet	apache-2.0	Gretel Synthetic Text-to-SQL dataset creators; source card and author citation preserved in the download manifest.	proposed	List the top 2 countries with the highest average artifact weight, along with the year and total weight of those artifacts.	CREATE TABLE ExcavationSites (SiteID INT, Country VARCHAR(50), Year INT, ArtifactWeight FLOAT); INSERT INTO ExcavationSites (SiteID, Country, Year, ArtifactWeight) VALUES (1, 'USA', 2020, 23.5), (2, 'Mexico', 2020, 14.2), (3, 'USA', 2019, 34.8), (4, 'Canada', 2019, 45.6), (5, 'Canada', 2019, 56.7);	database_business_operations	database_business_operations.database_query_analysis	heuristic_unreviewed	Mapped from source sql_task_type='analytics and reporting'; no individual SQL execution or semantic review.	en	none	SELECT Country, Year, SUM(ArtifactWeight) AS TotalWeight, AVG(ArtifactWeight) OVER (PARTITION BY Country) AS AvgWeight FROM (SELECT Country, Year, ArtifactWeight, ROW_NUMBER() OVER (PARTITION BY Country ORDER BY ArtifactWeight DESC) rn FROM ExcavationSites) x WHERE rn <= 2 GROUP BY Country, Year;	{"domain":"archeology","domain_description":"Detailed records on excavation sites, artifact analysis, historical context, and public outreach in archeology.","id":1321,"sql_complexity":"window functions","sql_complexity_description":"window functions (e.g., ROW_NUMBER, LEAD, LAG, RANk, NTILE, PERCENT_RANK, etc.) with p...	c696ef06f37dac90b076553b40e0c07a4a76d6aafa30e18a01af9122c2737db7	gretel_sql:test:1321	window functions	intermediate	Estimated from source sql_complexity='window functions'; query feature labels are only proxies for difficulty, not measured solver performance. No empirical solver calibration was performed.	heuristic	needs_difficulty_calibration	not_measured	analysis_only_unvalidated	primary	task_annotation	1,320	[ "difficulty_not_calibrated", "not_individually_reviewed", "reference_not_independently_verified", "reference_or_constraint_only_no_rubric", "response_diversity_not_validated", "source_evaluation_split", "sql_dialect_and_result_equivalence_need_review", "sql_not_executed", "synthetic_reference" ]	false	true	false	task:cb9973e895539e119850c9d0aa0f408b8d18a46b81d99ae55ea5b246f03b7253	1	1	[ "gretel_sql" ]	held_out	curated_candidate	[]	core_candidate	[ "source_sql_complexity:window functions", "sql_structure:selects=2;joins=0;statements=1;advanced=OVER", "combined_relational_operations_require_calibration" ]	structural_heuristic	Qwen3.6-27B	heuristic_not_model_measured
gretel_sql:e3ecd99a7bf2bd67b72a76d54c988e28c13ee5c918433fc1b23459c1acc48625	task:edc83842f8fafbd94b772ba78b614443a1ffe8907d292b2509d09048c19e99a6	e598fa238ed773468230da7058f1d7206f032a4700f66ebaec9338da3cf8801b	gretel_sql	gretelai/synthetic_text_to_sql	740ab236e64503fba51be1101df7a1be83bf455d	synthetic_text_to_sql_test.snappy.parquet	1367	test	https://huggingface.co/datasets/gretelai/synthetic_text_to_sql/resolve/740ab236e64503fba51be1101df7a1be83bf455d/synthetic_text_to_sql_test.snappy.parquet	apache-2.0	Gretel Synthetic Text-to-SQL dataset creators; source card and author citation preserved in the download manifest.	proposed	Calculate the percentage of vessels in each ocean basin that have outdated engine technology.	CREATE TABLE fleet_information (id INT, vessel_name VARCHAR(255), ocean_basin VARCHAR(255), engine_technology DATE); INSERT INTO fleet_information (id, vessel_name, ocean_basin, engine_technology) VALUES (1, 'Ocean Titan', 'Atlantic', '2000-01-01'), (2, 'Sea Explorer', 'Pacific', '2010-01-01');	database_business_operations	database_business_operations.database_query_analysis	heuristic_unreviewed	Mapped from source sql_task_type='analytics and reporting'; no individual SQL execution or semantic review.	en	none	SELECT ocean_basin, PERCENTAGE_RANK() OVER (ORDER BY outdated_engine_count) FROM (SELECT ocean_basin, COUNT(*) FILTER (WHERE engine_technology < '2010-01-01') AS outdated_engine_count FROM fleet_information GROUP BY ocean_basin);	{"domain":"oceanography","domain_description":"Marine life research data, ocean floor mapping projects, pollution control initiatives, and maritime law compliance.","id":1367,"sql_complexity":"window functions","sql_complexity_description":"window functions (e.g., ROW_NUMBER, LEAD, LAG, RANk, NTILE, PERCENT_RANK, etc.)...	c87e5117c200a782cabc93bf714049908f20ffa77eff37127faf7d58b5c2467b	gretel_sql:test:1367	window functions	intermediate	Estimated from source sql_complexity='window functions'; query feature labels are only proxies for difficulty, not measured solver performance. No empirical solver calibration was performed.	heuristic	needs_difficulty_calibration	not_measured	analysis_only_unvalidated	primary	task_annotation	1,366	[ "difficulty_not_calibrated", "not_individually_reviewed", "reference_not_independently_verified", "reference_or_constraint_only_no_rubric", "response_diversity_not_validated", "source_evaluation_split", "sql_dialect_and_result_equivalence_need_review", "sql_not_executed", "synthetic_reference" ]	false	true	false	task:edc83842f8fafbd94b772ba78b614443a1ffe8907d292b2509d09048c19e99a6	1	1	[ "gretel_sql" ]	held_out	curated_candidate	[]	core_candidate	[ "source_sql_complexity:window functions", "sql_structure:selects=2;joins=0;statements=1;advanced=OVER", "combined_relational_operations_require_calibration" ]	structural_heuristic	Qwen3.6-27B	heuristic_not_model_measured
gretel_sql:6a3cdfe57773484f1b9a61de4b9f559623407a67c90d1949985030e3c1f7631d	task:fc94554297817190d67058846e324366315728d4159d79425cf9ccee7d970a7e	8c7748ec80232bf76d6e89a057bddb591677b21f415a773de80288f583b5927d	gretel_sql	gretelai/synthetic_text_to_sql	740ab236e64503fba51be1101df7a1be83bf455d	synthetic_text_to_sql_test.snappy.parquet	1510	test	https://huggingface.co/datasets/gretelai/synthetic_text_to_sql/resolve/740ab236e64503fba51be1101df7a1be83bf455d/synthetic_text_to_sql_test.snappy.parquet	apache-2.0	Gretel Synthetic Text-to-SQL dataset creators; source card and author citation preserved in the download manifest.	proposed	What is the total number of visitors from African countries who attended events in 2021?	CREATE TABLE events (event_id INT, event_name VARCHAR(50), event_year INT, location VARCHAR(50)); INSERT INTO events (event_id, event_name, event_year, location) VALUES (1, 'Music Festival', 2021, 'Nigeria'), (2, 'Art Exhibition', 2022, 'Egypt'), (3, 'Theater Performance', 2021, 'South Africa'); CREATE TABLE countries ...	database_business_operations	database_business_operations.database_query_analysis	heuristic_unreviewed	Mapped from source sql_task_type='analytics and reporting'; no individual SQL execution or semantic review.	en	none	SELECT COUNT(*) FROM attendees JOIN events ON attendees.event_id = events.event_id JOIN countries ON attendees.country_id = countries.country_id WHERE events.event_year = 2021 AND countries.continent = 'Africa';	{"domain":"arts and culture","domain_description":"Audience demographics, event attendance, program impact, and funding sources.","id":1510,"sql_complexity":"multiple_joins","sql_complexity_description":"two or more joins (specify inner, outer, cross)","sql_explanation":"First, we join the attendees, events, and countr...	061f9414ff6fa38423e0203736d47a4b0a4f8dbc180b7088fb672d799f2fce71	gretel_sql:test:1510	multiple_joins	intermediate	Estimated from source sql_complexity='multiple_joins'; query feature labels are only proxies for difficulty, not measured solver performance. No empirical solver calibration was performed.	heuristic	needs_difficulty_calibration	not_measured	analysis_only_unvalidated	primary	task_annotation	1,509	[ "difficulty_not_calibrated", "not_individually_reviewed", "reference_not_independently_verified", "reference_or_constraint_only_no_rubric", "response_diversity_not_validated", "source_evaluation_split", "sql_dialect_and_result_equivalence_need_review", "sql_not_executed", "synthetic_reference" ]	false	true	false	task:fc94554297817190d67058846e324366315728d4159d79425cf9ccee7d970a7e	1	1	[ "gretel_sql" ]	held_out	curated_candidate	[]	core_candidate	[ "source_sql_complexity:multiple_joins", "sql_structure:selects=1;joins=2;statements=1;advanced=none", "combined_relational_operations_require_calibration" ]	structural_heuristic	Qwen3.6-27B	heuristic_not_model_measured
gretel_sql:de6d37a92c3ff155f2d04d585285955980637b82bef19a0b350ca72c3347b1b2	task:6a323d2babdc9b885be5d156e2f009b6ce8e0c26ea3a6b36c1b4f23326e939be	a07a6258eb5cd19a0234ac344221d4fe0c7624bd7538fd4239fe552877964167	gretel_sql	gretelai/synthetic_text_to_sql	740ab236e64503fba51be1101df7a1be83bf455d	synthetic_text_to_sql_test.snappy.parquet	1515	test	https://huggingface.co/datasets/gretelai/synthetic_text_to_sql/resolve/740ab236e64503fba51be1101df7a1be83bf455d/synthetic_text_to_sql_test.snappy.parquet	apache-2.0	Gretel Synthetic Text-to-SQL dataset creators; source card and author citation preserved in the download manifest.	proposed	What is the average age of clients who lost cases in the 'personal injury' category?	CREATE TABLE Cases (CaseID int, ClientID int, Category varchar(50)); INSERT INTO Cases (CaseID, ClientID, Category) VALUES (701, 7, 'Personal Injury'); CREATE TABLE Clients (ClientID int, Age int, Gender varchar(10)); INSERT INTO Clients (ClientID, Age, Gender) VALUES (7, 45, 'Male'); CREATE TABLE CaseOutcomes (CaseID ...	database_business_operations	database_business_operations.database_query_analysis	heuristic_unreviewed	Mapped from source sql_task_type='analytics and reporting'; no individual SQL execution or semantic review.	en	none	SELECT AVG(C.Age) as AvgAge FROM Clients C INNER JOIN Cases CA ON C.ClientID = CA.ClientID INNER JOIN CaseOutcomes CO ON CA.CaseID = CO.CaseID WHERE CA.Category = 'Personal Injury' AND CO.Outcome = 'Lost';	{"domain":"legal services","domain_description":"Case outcomes, legal precedents, attorney performance metrics, client demographics, and billing information.","id":1515,"sql_complexity":"multiple_joins","sql_complexity_description":"two or more joins (specify inner, outer, cross)","sql_explanation":"This query finds th...	45e914721c988f5409960a8d80da4230fce881d47ee66a0c493a3d7c2a1fb4ac	gretel_sql:test:1515	multiple_joins	intermediate	Estimated from source sql_complexity='multiple_joins'; query feature labels are only proxies for difficulty, not measured solver performance. No empirical solver calibration was performed.	heuristic	needs_difficulty_calibration	not_measured	analysis_only_unvalidated	primary	task_annotation	1,514	[ "difficulty_not_calibrated", "not_individually_reviewed", "reference_not_independently_verified", "reference_or_constraint_only_no_rubric", "response_diversity_not_validated", "source_evaluation_split", "sql_dialect_and_result_equivalence_need_review", "sql_not_executed", "synthetic_reference" ]	false	true	false	task:6a323d2babdc9b885be5d156e2f009b6ce8e0c26ea3a6b36c1b4f23326e939be	1	1	[ "gretel_sql" ]	held_out	curated_candidate	[]	core_candidate	[ "source_sql_complexity:multiple_joins", "sql_structure:selects=1;joins=2;statements=1;advanced=none", "combined_relational_operations_require_calibration" ]	structural_heuristic	Qwen3.6-27B	heuristic_not_model_measured
gretel_sql:ef79c30a0789531703b8eddb68073b5375da80666becd02149f23ff2c0031af4	task:26fc01a2e492f7de5be58c73ca71b71bf70c516071ede4b4acea04ff6992a246	a3622d5b31cb16418d3e1e521bfd13eb14ee998344830a548993cc6fd511c039	gretel_sql	gretelai/synthetic_text_to_sql	740ab236e64503fba51be1101df7a1be83bf455d	synthetic_text_to_sql_test.snappy.parquet	1531	test	https://huggingface.co/datasets/gretelai/synthetic_text_to_sql/resolve/740ab236e64503fba51be1101df7a1be83bf455d/synthetic_text_to_sql_test.snappy.parquet	apache-2.0	Gretel Synthetic Text-to-SQL dataset creators; source card and author citation preserved in the download manifest.	proposed	How many transactions were made in each region for the 'Credit Cards' product type?	CREATE TABLE regions (id INT, region_name VARCHAR(50)); INSERT INTO regions (id, region_name) VALUES (1, 'Northeast'), (2, 'Southeast'); CREATE TABLE transactions (region_id INT, product_type_id INT, transaction_count INT); INSERT INTO transactions (region_id, product_type_id, transaction_count) VALUES (1, 1, 20), (1, ...	database_business_operations	database_business_operations.database_query_analysis	heuristic_unreviewed	Mapped from source sql_task_type='analytics and reporting'; no individual SQL execution or semantic review.	en	none	SELECT r.region_name, p.product_type, SUM(t.transaction_count) as total_transactions FROM regions r JOIN transactions t ON r.id = t.region_id JOIN product_types p ON t.product_type_id = p.id WHERE p.product_type = 'Credit Cards' GROUP BY r.region_name, p.product_type;	{"domain":"financial services","domain_description":"Detailed financial data including investment strategies, risk management, fraud detection, customer analytics, and regulatory compliance.","id":1531,"sql_complexity":"multiple_joins","sql_complexity_description":"two or more joins (specify inner, outer, cross)","sql_...	0d900449eb2a652885116c659921c5b31cb24ea676531ef7b767aa33ea3be555	gretel_sql:test:1531	multiple_joins	intermediate	Estimated from source sql_complexity='multiple_joins'; query feature labels are only proxies for difficulty, not measured solver performance. No empirical solver calibration was performed.	heuristic	needs_difficulty_calibration	not_measured	analysis_only_unvalidated	primary	task_annotation	1,530	[ "difficulty_not_calibrated", "not_individually_reviewed", "reference_not_independently_verified", "reference_or_constraint_only_no_rubric", "response_diversity_not_validated", "source_evaluation_split", "sql_dialect_and_result_equivalence_need_review", "sql_not_executed", "synthetic_reference" ]	false	true	false	task:26fc01a2e492f7de5be58c73ca71b71bf70c516071ede4b4acea04ff6992a246	1	1	[ "gretel_sql" ]	held_out	curated_candidate	[]	core_candidate	[ "source_sql_complexity:multiple_joins", "sql_structure:selects=1;joins=2;statements=1;advanced=none", "combined_relational_operations_require_calibration" ]	structural_heuristic	Qwen3.6-27B	heuristic_not_model_measured
gretel_sql:a03e79bd8209d499ff19973e14231b79b4cb1545a9d064ad5a0d971e4cb433fc	task:f7b1bf27656cb27292cf90eb7fef0ee661651d5c35ae8011ea319b3c49a38ae0	781fcdffd244553d8701b9d474f72b3dae1f423888bc4cbf721b38ee331358eb	gretel_sql	gretelai/synthetic_text_to_sql	740ab236e64503fba51be1101df7a1be83bf455d	synthetic_text_to_sql_test.snappy.parquet	1562	test	https://huggingface.co/datasets/gretelai/synthetic_text_to_sql/resolve/740ab236e64503fba51be1101df7a1be83bf455d/synthetic_text_to_sql_test.snappy.parquet	apache-2.0	Gretel Synthetic Text-to-SQL dataset creators; source card and author citation preserved in the download manifest.	proposed	What is the total CO2 emission of each menu item, considering its ingredients and their origin?	CREATE TABLE menu_items (menu_id INT, name VARCHAR(50), co2_emission FLOAT); CREATE TABLE ingredients (ingredient_id INT, name VARCHAR(50), origin VARCHAR(50), co2_emission_per_kg FLOAT); CREATE TABLE recipe (menu_id INT, ingredient_id INT, quantity FLOAT);	database_business_operations	database_business_operations.database_query_analysis	heuristic_unreviewed	Mapped from source sql_task_type='analytics and reporting'; no individual SQL execution or semantic review.	en	none	SELECT m.name, SUM(i.co2_emission_per_kg * r.quantity) as total_co2_emission FROM menu_items m JOIN recipe r ON m.menu_id = r.menu_id JOIN ingredients i ON r.ingredient_id = i.ingredient_id GROUP BY m.menu_id;	{"domain":"food services","domain_description":"Menu engineering, customer preferences, inventory management, and sustainability initiatives.","id":1562,"sql_complexity":"multiple_joins","sql_complexity_description":"two or more joins (specify inner, outer, cross)","sql_explanation":"The SQL query joins the menu_items,...	1875055bfcec78d42960e73e1923895715a1fa5cf6b4cae67be64e711363c2d5	gretel_sql:test:1562	multiple_joins	intermediate	Estimated from source sql_complexity='multiple_joins'; query feature labels are only proxies for difficulty, not measured solver performance. No empirical solver calibration was performed.	heuristic	needs_difficulty_calibration	not_measured	analysis_only_unvalidated	primary	task_annotation	1,561	[ "difficulty_not_calibrated", "not_individually_reviewed", "reference_not_independently_verified", "reference_or_constraint_only_no_rubric", "response_diversity_not_validated", "source_evaluation_split", "sql_dialect_and_result_equivalence_need_review", "sql_not_executed", "synthetic_reference" ]	false	true	false	task:f7b1bf27656cb27292cf90eb7fef0ee661651d5c35ae8011ea319b3c49a38ae0	1	1	[ "gretel_sql" ]	held_out	curated_candidate	[]	core_candidate	[ "source_sql_complexity:multiple_joins", "sql_structure:selects=1;joins=2;statements=1;advanced=none", "combined_relational_operations_require_calibration" ]	structural_heuristic	Qwen3.6-27B	heuristic_not_model_measured
gretel_sql:17792feb0a5139588e62172c1cc3bb87e05d71b8d776d64aa740dabcfca85719	task:7a2c93c498aa9fdb7b1071cd589756b310e45021cedd69eed22e6bb1e28c4a80	4538afbbbc45752bc61243c62d346e948e662f07752ca6e98c59afd84ef322d4	gretel_sql	gretelai/synthetic_text_to_sql	740ab236e64503fba51be1101df7a1be83bf455d	synthetic_text_to_sql_test.snappy.parquet	1584	test	https://huggingface.co/datasets/gretelai/synthetic_text_to_sql/resolve/740ab236e64503fba51be1101df7a1be83bf455d/synthetic_text_to_sql_test.snappy.parquet	apache-2.0	Gretel Synthetic Text-to-SQL dataset creators; source card and author citation preserved in the download manifest.	proposed	What are the names of the vessels with the highest average speed that arrived in Busan?	CREATE TABLE VesselArrivals (ID INT, VesselName VARCHAR(50), ArrivalPort VARCHAR(50), ArrivalDate DATE, AverageSpeed DECIMAL(5,2)); INSERT INTO VesselArrivals (ID, VesselName, ArrivalPort, ArrivalDate, AverageSpeed) VALUES (1, 'Test Vessel 1', 'Busan', '2022-01-01', 20.0), (2, 'Test Vessel 2', 'Busan', '2022-01-02', 18...	database_business_operations	database_business_operations.database_query_analysis	heuristic_unreviewed	Mapped from source sql_task_type='analytics and reporting'; no individual SQL execution or semantic review.	en	none	SELECT VesselName FROM (SELECT VesselName, ROW_NUMBER() OVER (ORDER BY AverageSpeed DESC) AS rn FROM VesselArrivals WHERE ArrivalPort = 'Busan') t WHERE rn = 1;	{"domain":"maritime","domain_description":"Vessel performance data, cargo tracking, safety records, and regulatory compliance.","id":1584,"sql_complexity":"window functions","sql_complexity_description":"window functions (e.g., ROW_NUMBER, LEAD, LAG, RANk, NTILE, PERCENT_RANK, etc.) with partitioning and ordering","sql...	9122ca8044d6a310183df14adb53a14832637af95426a3b45ddf0793998b4e5e	gretel_sql:test:1584	window functions	intermediate	Estimated from source sql_complexity='window functions'; query feature labels are only proxies for difficulty, not measured solver performance. No empirical solver calibration was performed.	heuristic	needs_difficulty_calibration	not_measured	analysis_only_unvalidated	primary	task_annotation	1,583	[ "difficulty_not_calibrated", "not_individually_reviewed", "reference_not_independently_verified", "reference_or_constraint_only_no_rubric", "response_diversity_not_validated", "source_evaluation_split", "sql_dialect_and_result_equivalence_need_review", "sql_not_executed", "synthetic_reference" ]	false	true	false	task:7a2c93c498aa9fdb7b1071cd589756b310e45021cedd69eed22e6bb1e28c4a80	1	1	[ "gretel_sql" ]	held_out	curated_candidate	[]	core_candidate	[ "source_sql_complexity:window functions", "sql_structure:selects=2;joins=0;statements=1;advanced=OVER", "combined_relational_operations_require_calibration" ]	structural_heuristic	Qwen3.6-27B	heuristic_not_model_measured
gretel_sql:e328b528b1500e06131037295b1af701c8e1421c8caf806677f5f55e13b239cb	task:d99418893fc67eab0e4011e5aad6940506d35ae6fff3405b5b08e8d3313cd2a4	556ad253c9b5b9759d9a323fb0e22c66f04ee6c1f7373200051e5d83834d771a	gretel_sql	gretelai/synthetic_text_to_sql	740ab236e64503fba51be1101df7a1be83bf455d	synthetic_text_to_sql_test.snappy.parquet	1610	test	https://huggingface.co/datasets/gretelai/synthetic_text_to_sql/resolve/740ab236e64503fba51be1101df7a1be83bf455d/synthetic_text_to_sql_test.snappy.parquet	apache-2.0	Gretel Synthetic Text-to-SQL dataset creators; source card and author citation preserved in the download manifest.	proposed	What is the number of primary care physicians per capita in each state of the United States?	CREATE TABLE us_states (id INT, name VARCHAR(255)); CREATE TABLE primary_care_physicians (id INT, state_id INT, count INT); CREATE TABLE population (id INT, state_id INT, total_population INT); INSERT INTO us_states (id, name) VALUES (1, 'Alabama'), (2, 'Alaska'), (3, 'Arizona'), (4, 'Arkansas'), (5, 'California');	database_business_operations	database_business_operations.database_query_analysis	heuristic_unreviewed	Mapped from source sql_task_type='analytics and reporting'; no individual SQL execution or semantic review.	en	none	SELECT s.name, pc.count / p.total_population AS physicians_per_capita FROM primary_care_physicians pc JOIN us_states s ON pc.state_id = s.id JOIN population p ON pc.state_id = p.state_id;	{"domain":"public health","domain_description":"Community health statistics, infectious disease tracking data, healthcare access metrics, and public health policy analysis.","id":1610,"sql_complexity":"multiple_joins","sql_complexity_description":"two or more joins (specify inner, outer, cross)","sql_explanation":"This...	e8a5afbee2ce4c58702ebfadfd9e358a70490e21eaeaf43c480eb738b2b47993	gretel_sql:test:1610	multiple_joins	intermediate	Estimated from source sql_complexity='multiple_joins'; query feature labels are only proxies for difficulty, not measured solver performance. No empirical solver calibration was performed.	heuristic	needs_difficulty_calibration	not_measured	analysis_only_unvalidated	primary	task_annotation	1,609	[ "difficulty_not_calibrated", "not_individually_reviewed", "reference_not_independently_verified", "reference_or_constraint_only_no_rubric", "response_diversity_not_validated", "source_evaluation_split", "sql_dialect_and_result_equivalence_need_review", "sql_not_executed", "synthetic_reference" ]	false	true	false	task:d99418893fc67eab0e4011e5aad6940506d35ae6fff3405b5b08e8d3313cd2a4	1	1	[ "gretel_sql" ]	held_out	curated_candidate	[]	core_candidate	[ "source_sql_complexity:multiple_joins", "sql_structure:selects=1;joins=2;statements=1;advanced=none", "combined_relational_operations_require_calibration" ]	structural_heuristic	Qwen3.6-27B	heuristic_not_model_measured
gretel_sql:04eb006484af290c371699dbd4c5c4dc8954750e1c41efad7a57d89e158d076b	task:fbaca957d2329d4181b84f5712d9d34a05ed8780473579ad60ded5acaa1a500d	ef9a5e5718b0f1d4f4abe5d1e1c9682c5b719cb8092d6816fb9ad416607caa52	gretel_sql	gretelai/synthetic_text_to_sql	740ab236e64503fba51be1101df7a1be83bf455d	synthetic_text_to_sql_test.snappy.parquet	1669	test	https://huggingface.co/datasets/gretelai/synthetic_text_to_sql/resolve/740ab236e64503fba51be1101df7a1be83bf455d/synthetic_text_to_sql_test.snappy.parquet	apache-2.0	Gretel Synthetic Text-to-SQL dataset creators; source card and author citation preserved in the download manifest.	proposed	Which countries have the highest and lowest ocean acidity levels?	CREATE TABLE ocean_acidity (country TEXT, avg_ph REAL); INSERT INTO ocean_acidity (country, avg_ph) VALUES ('United States', 7.8), ('Canada', 7.6), ('Mexico', 7.9);	database_business_operations	database_business_operations.database_query_analysis	heuristic_unreviewed	Mapped from source sql_task_type='analytics and reporting'; no individual SQL execution or semantic review.	en	none	SELECT country, avg_ph FROM (SELECT country, avg_ph, ROW_NUMBER() OVER (ORDER BY avg_ph DESC) as rnk FROM ocean_acidity) subq WHERE rnk = 1 OR rnk = (SELECT COUNT(*) FROM ocean_acidity) ORDER BY avg_ph;	{"domain":"oceans","domain_description":"Ocean data on marine conservation, ocean acidification, deep-sea exploration, and maritime safety.","id":1669,"sql_complexity":"window functions","sql_complexity_description":"window functions (e.g., ROW_NUMBER, LEAD, LAG, RANk, NTILE, PERCENT_RANK, etc.) with partitioning and o...	4f013a3cdceba239d4648392d0c92cbe6d0d21ceac5864c026e423f894450e39	gretel_sql:test:1669	window functions	intermediate	Estimated from source sql_complexity='window functions'; query feature labels are only proxies for difficulty, not measured solver performance. No empirical solver calibration was performed.	heuristic	needs_difficulty_calibration	not_measured	analysis_only_unvalidated	primary	task_annotation	1,668	[ "difficulty_not_calibrated", "not_individually_reviewed", "reference_not_independently_verified", "reference_or_constraint_only_no_rubric", "response_diversity_not_validated", "source_evaluation_split", "sql_dialect_and_result_equivalence_need_review", "sql_not_executed", "synthetic_reference" ]	false	true	false	task:fbaca957d2329d4181b84f5712d9d34a05ed8780473579ad60ded5acaa1a500d	1	1	[ "gretel_sql" ]	held_out	curated_candidate	[]	core_candidate	[ "source_sql_complexity:window functions", "sql_structure:selects=3;joins=0;statements=1;advanced=OVER", "combined_relational_operations_require_calibration" ]	structural_heuristic	Qwen3.6-27B	heuristic_not_model_measured
gretel_sql:c0e478e277636d4ded47904f433733506cfa4cb5de5a26d9a5980e2af4d6804d	task:1cbd92181b981f0d989be02ec431a7c59a3efc3aa3ada1a26370b30542ee757a	b362c64856db8b1a86006db6e086da870e4a79ba72cb4bb88ff89994a38e60f2	gretel_sql	gretelai/synthetic_text_to_sql	740ab236e64503fba51be1101df7a1be83bf455d	synthetic_text_to_sql_test.snappy.parquet	1672	test	https://huggingface.co/datasets/gretelai/synthetic_text_to_sql/resolve/740ab236e64503fba51be1101df7a1be83bf455d/synthetic_text_to_sql_test.snappy.parquet	apache-2.0	Gretel Synthetic Text-to-SQL dataset creators; source card and author citation preserved in the download manifest.	proposed	Identify the top 3 countries with the highest number of players who have adopted virtual reality technology.	CREATE TABLE CountryData (Country VARCHAR(50), Population INT, VRAdopters INT); INSERT INTO CountryData (Country, Population, VRAdopters) VALUES ('USA', 331002651, 50000), ('China', 1439323776, 25000), ('Canada', 37410003), ('India', 1380004385, 10000);	database_business_operations	database_business_operations.database_query_analysis	heuristic_unreviewed	Mapped from source sql_task_type='analytics and reporting'; no individual SQL execution or semantic review.	en	none	SELECT Country, VRAdopters FROM (SELECT Country, VRAdopters, ROW_NUMBER() OVER (ORDER BY VRAdopters DESC) AS RN FROM CountryData JOIN (SELECT PlayerID, VRDevice FROM VRAdoption GROUP BY PlayerID, VRDevice) VR ON CountryData.Country = VR.PlayerCountry) T WHERE RN <= 3	{"domain":"gaming technology","domain_description":"Player demographics, game design data, esports event information, and virtual reality technology adoption.","id":1672,"sql_complexity":"window functions","sql_complexity_description":"window functions (e.g., ROW_NUMBER, LEAD, LAG, RANk, NTILE, PERCENT_RANK, etc.) with...	3068b9cd7ccf2b89247cb76954700b5f80b5bf7d549feefcb041ddeee45de519	gretel_sql:test:1672	window functions	intermediate	Estimated from source sql_complexity='window functions'; query feature labels are only proxies for difficulty, not measured solver performance. No empirical solver calibration was performed.	heuristic	needs_difficulty_calibration	not_measured	analysis_only_unvalidated	primary	task_annotation	1,671	[ "difficulty_not_calibrated", "not_individually_reviewed", "reference_not_independently_verified", "reference_or_constraint_only_no_rubric", "response_diversity_not_validated", "source_evaluation_split", "sql_dialect_and_result_equivalence_need_review", "sql_not_executed", "synthetic_reference" ]	false	true	false	task:1cbd92181b981f0d989be02ec431a7c59a3efc3aa3ada1a26370b30542ee757a	1	1	[ "gretel_sql" ]	held_out	curated_candidate	[]	core_candidate	[ "source_sql_complexity:window functions", "sql_structure:selects=3;joins=1;statements=1;advanced=OVER", "combined_relational_operations_require_calibration" ]	structural_heuristic	Qwen3.6-27B	heuristic_not_model_measured
gretel_sql:e6d059f342cfb9d93e11ad20293db21f723c31571a68cd472283aad509d429a7	task:fd78f98f90969e07e25286eef9e2323a4edc8a80b1c7effc65a247928fa03caf	1ad1ade9e9912e3fe9761be4c2cbd7bcfe918b3ebff5006af68afe8f8eaedd63	gretel_sql	gretelai/synthetic_text_to_sql	740ab236e64503fba51be1101df7a1be83bf455d	synthetic_text_to_sql_test.snappy.parquet	1751	test	https://huggingface.co/datasets/gretelai/synthetic_text_to_sql/resolve/740ab236e64503fba51be1101df7a1be83bf455d/synthetic_text_to_sql_test.snappy.parquet	apache-2.0	Gretel Synthetic Text-to-SQL dataset creators; source card and author citation preserved in the download manifest.	proposed	What is the maximum consecutive number of days with a financial wellbeing score below 60 for each customer?	CREATE TABLE customer_scores (customer_id INT, score_date DATE, financial_wellbeing_score INT); INSERT INTO customer_scores (customer_id, score_date, financial_wellbeing_score) VALUES (1, '2021-01-01', 65), (1, '2021-01-02', 60), (1, '2021-01-03', 55), (1, '2021-01-04', 60), (2, '2021-01-01', 70), (2, '2021-01-02', 75)...	database_business_operations	database_business_operations.database_query_analysis	heuristic_unreviewed	Mapped from source sql_task_type='analytics and reporting'; no individual SQL execution or semantic review.	en	none	SELECT customer_id, MAX(consecutive_below_60) FROM (SELECT customer_id, score_date, financial_wellbeing_score, COUNT(*) FILTER (WHERE financial_wellbeing_score < 60) OVER (PARTITION BY customer_id ORDER BY score_date ROWS BETWEEN UNBOUNDED PRECEDING AND 1 PRECEDING) AS consecutive_below_60 FROM customer_scores) subquer...	{"domain":"finance","domain_description":"Financial data on Shariah-compliant finance, socially responsible lending, financial capability, and financial wellbeing.","id":1751,"sql_complexity":"window functions","sql_complexity_description":"window functions (e.g., ROW_NUMBER, LEAD, LAG, RANk, NTILE, PERCENT_RANK, etc.)...	d1919d3b018e586512741aa56bc453592d712354910f360d2c0a55fab447abb4	gretel_sql:test:1751	window functions	intermediate	Estimated from source sql_complexity='window functions'; query feature labels are only proxies for difficulty, not measured solver performance. No empirical solver calibration was performed.	heuristic	needs_difficulty_calibration	not_measured	analysis_only_unvalidated	primary	task_annotation	1,750	[ "difficulty_not_calibrated", "not_individually_reviewed", "reference_not_independently_verified", "reference_or_constraint_only_no_rubric", "response_diversity_not_validated", "source_evaluation_split", "sql_dialect_and_result_equivalence_need_review", "sql_not_executed", "synthetic_reference" ]	false	true	false	task:fd78f98f90969e07e25286eef9e2323a4edc8a80b1c7effc65a247928fa03caf	1	1	[ "gretel_sql" ]	held_out	curated_candidate	[]	core_candidate	[ "source_sql_complexity:window functions", "sql_structure:selects=2;joins=0;statements=1;advanced=OVER", "combined_relational_operations_require_calibration" ]	structural_heuristic	Qwen3.6-27B	heuristic_not_model_measured
gretel_sql:cde37366c9a77951ab03d4af9e2e0a8cb805cf3a54c0f932e333137b1870538e	task:fcaa3dc216b5f2edc188454a81e45b073278a6294b32f28ab7636896e40ccd40	e8d13bf1b649c99ca3737287e6957bb2fefdd4c1374842a512c0fa1c234b11c3	gretel_sql	gretelai/synthetic_text_to_sql	740ab236e64503fba51be1101df7a1be83bf455d	synthetic_text_to_sql_test.snappy.parquet	1763	test	https://huggingface.co/datasets/gretelai/synthetic_text_to_sql/resolve/740ab236e64503fba51be1101df7a1be83bf455d/synthetic_text_to_sql_test.snappy.parquet	apache-2.0	Gretel Synthetic Text-to-SQL dataset creators; source card and author citation preserved in the download manifest.	proposed	What is the total food and beverage revenue last month for hotels in 'Bangkok'?	CREATE TABLE revenue (hotel_id INT, revenue_source VARCHAR(50), revenue INT, revenue_date DATE); INSERT INTO revenue (hotel_id, revenue_source, revenue, revenue_date) VALUES (5, 'Room revenue', 12000, '2022-03-01'), (5, 'Food and beverage', 4000, '2022-03-02'), (5, 'Other revenue', 1000, '2022-03-03'); CREATE TABLE hot...	database_business_operations	database_business_operations.database_query_analysis	heuristic_unreviewed	Mapped from source sql_task_type='analytics and reporting'; no individual SQL execution or semantic review.	en	none	SELECT SUM(revenue) FROM revenue JOIN hotels ON revenue.hotel_id = hotels.hotel_id JOIN dates ON revenue.revenue_date = dates.date WHERE hotels.city = 'Bangkok' AND revenue_source = 'Food and beverage' AND dates.date >= DATEADD(month, -1, GETDATE());	{"domain":"hospitality technology","domain_description":"Hotel tech adoption metrics, online travel agency data, virtual tour engagement stats, and hospitality AI trends.","id":1763,"sql_complexity":"multiple_joins","sql_complexity_description":"two or more joins (specify inner, outer, cross)","sql_explanation":"The SQ...	7523c6adc7cb6d6bbfe300c5712a1a2b08b05b868db61795f894fcb1ef13b685	gretel_sql:test:1763	multiple_joins	intermediate	Estimated from source sql_complexity='multiple_joins'; query feature labels are only proxies for difficulty, not measured solver performance. No empirical solver calibration was performed.	heuristic	needs_difficulty_calibration	not_measured	analysis_only_unvalidated	primary	task_annotation	1,762	[ "difficulty_not_calibrated", "not_individually_reviewed", "reference_not_independently_verified", "reference_or_constraint_only_no_rubric", "response_diversity_not_validated", "source_evaluation_split", "sql_dialect_and_result_equivalence_need_review", "sql_not_executed", "synthetic_reference" ]	false	true	false	task:fcaa3dc216b5f2edc188454a81e45b073278a6294b32f28ab7636896e40ccd40	1	1	[ "gretel_sql" ]	held_out	curated_candidate	[]	core_candidate	[ "source_sql_complexity:multiple_joins", "sql_structure:selects=1;joins=2;statements=1;advanced=none", "combined_relational_operations_require_calibration" ]	structural_heuristic	Qwen3.6-27B	heuristic_not_model_measured
gretel_sql:7415f71484ee09a985eeb1d31c7e8d19c8b07d102b6b4677672b4cb6c2c34639	task:dcc326007f141362f4eaabfceae77dc9ddd0ce8e671ea89f4d4b62d291b43ff3	bf53e2f4d721b2d5ff8effdd7b6b6cac8e18fad6caf5df9c3c890f6bf645b90c	gretel_sql	gretelai/synthetic_text_to_sql	740ab236e64503fba51be1101df7a1be83bf455d	synthetic_text_to_sql_test.snappy.parquet	1785	test	https://huggingface.co/datasets/gretelai/synthetic_text_to_sql/resolve/740ab236e64503fba51be1101df7a1be83bf455d/synthetic_text_to_sql_test.snappy.parquet	apache-2.0	Gretel Synthetic Text-to-SQL dataset creators; source card and author citation preserved in the download manifest.	proposed	Who are the top 5 vendors with the highest revenue from circular supply chain products?	CREATE TABLE vendors (vendor_id INT, name TEXT); CREATE TABLE sales (sale_id INT, vendor_id INT, product_id INT, price DECIMAL(5,2)); INSERT INTO vendors (vendor_id, name) VALUES (1, 'Vendor A'), (2, 'Vendor B'), (3, 'Vendor C'), (4, 'Vendor D'), (5, 'Vendor E'); INSERT INTO sales (sale_id, vendor_id, product_id, price...	database_business_operations	database_business_operations.database_query_analysis	heuristic_unreviewed	Mapped from source sql_task_type='analytics and reporting'; no individual SQL execution or semantic review.	en	none	SELECT vendors.name, SUM(sales.price) FROM vendors INNER JOIN sales ON vendors.vendor_id = sales.vendor_id INNER JOIN circular_supply_chain_products ON sales.product_id = circular_supply_chain_products.product_id GROUP BY vendors.name ORDER BY SUM(sales.price) DESC LIMIT 5;	{"domain":"retail","domain_description":"Retail data on circular supply chains, ethical labor practices, product transparency, and consumer education.","id":1785,"sql_complexity":"multiple_joins","sql_complexity_description":"two or more joins (specify inner, outer, cross)","sql_explanation":"This query identifies the ...	fc8ccd031e3a27b203169f548f5083d0be5b734500d372a9644db21f7d7087f9	gretel_sql:test:1785	multiple_joins	intermediate	Estimated from source sql_complexity='multiple_joins'; query feature labels are only proxies for difficulty, not measured solver performance. No empirical solver calibration was performed.	heuristic	needs_difficulty_calibration	not_measured	analysis_only_unvalidated	primary	task_annotation	1,784	[ "difficulty_not_calibrated", "not_individually_reviewed", "reference_not_independently_verified", "reference_or_constraint_only_no_rubric", "response_diversity_not_validated", "source_evaluation_split", "sql_dialect_and_result_equivalence_need_review", "sql_not_executed", "synthetic_reference" ]	false	true	false	task:dcc326007f141362f4eaabfceae77dc9ddd0ce8e671ea89f4d4b62d291b43ff3	1	1	[ "gretel_sql" ]	held_out	curated_candidate	[]	core_candidate	[ "source_sql_complexity:multiple_joins", "sql_structure:selects=1;joins=2;statements=1;advanced=none", "combined_relational_operations_require_calibration" ]	structural_heuristic	Qwen3.6-27B	heuristic_not_model_measured
gretel_sql:0da0664771d1eaf5015177ae1b3d73052fb8e773d7b9578164af13f60592b0e0	task:ec32a2756dcb019d2f56302ee7ec7ef7ddae449652e9a058e654e4623191feab	be874cdf27e189ec588305d76aeedf12fb79cc38c518926a486aaa70e730a13d	gretel_sql	gretelai/synthetic_text_to_sql	740ab236e64503fba51be1101df7a1be83bf455d	synthetic_text_to_sql_test.snappy.parquet	1819	test	https://huggingface.co/datasets/gretelai/synthetic_text_to_sql/resolve/740ab236e64503fba51be1101df7a1be83bf455d/synthetic_text_to_sql_test.snappy.parquet	apache-2.0	Gretel Synthetic Text-to-SQL dataset creators; source card and author citation preserved in the download manifest.	proposed	Find the top 3 longest songs in the Rock genre.	CREATE TABLE songs (song_id INT, song_length FLOAT, genre TEXT); INSERT INTO songs VALUES (1, 450.3, 'Rock'), (2, 320.2, 'Pop'), (3, 500.5, 'Rock'), (4, 200.1, 'Jazz'), (5, 400.0, 'Rock');	database_business_operations	database_business_operations.database_query_analysis	heuristic_unreviewed	Mapped from source sql_task_type='analytics and reporting'; no individual SQL execution or semantic review.	en	none	SELECT * FROM (SELECT song_id, song_length, genre, ROW_NUMBER() OVER (ORDER BY song_length DESC) AS row_num FROM songs WHERE genre = 'Rock') AS subquery WHERE row_num <= 3;	{"domain":"music","domain_description":"Detailed records on music creation, distribution, marketing, and consumption patterns across various genres and platforms.","id":1819,"sql_complexity":"window functions","sql_complexity_description":"window functions (e.g., ROW_NUMBER, LEAD, LAG, RANk, NTILE, PERCENT_RANK, etc.) ...	e3e4d51c15172cd8a9f97ee0f08bc950f6f5c62cc38ac0e66d59349d6fb94264	gretel_sql:test:1819	window functions	intermediate	Estimated from source sql_complexity='window functions'; query feature labels are only proxies for difficulty, not measured solver performance. No empirical solver calibration was performed.	heuristic	needs_difficulty_calibration	not_measured	analysis_only_unvalidated	primary	task_annotation	1,818	[ "difficulty_not_calibrated", "not_individually_reviewed", "reference_not_independently_verified", "reference_or_constraint_only_no_rubric", "response_diversity_not_validated", "source_evaluation_split", "sql_dialect_and_result_equivalence_need_review", "sql_not_executed", "synthetic_reference" ]	false	true	false	task:ec32a2756dcb019d2f56302ee7ec7ef7ddae449652e9a058e654e4623191feab	1	1	[ "gretel_sql" ]	held_out	curated_candidate	[]	core_candidate	[ "source_sql_complexity:window functions", "sql_structure:selects=2;joins=0;statements=1;advanced=OVER", "combined_relational_operations_require_calibration" ]	structural_heuristic	Qwen3.6-27B	heuristic_not_model_measured
gretel_sql:361c533651390b2038a3771d3529cc69edafa52af7c1c7d5ff672470c491da81	task:1ba7172f82a48399b6904ad5d93cd163e76720bafa1ebd6f19d7ff5d2fd6ad31	9b922197758ed21697d12000c60a339cd73852e207881fca8288b69e27a46dea	gretel_sql	gretelai/synthetic_text_to_sql	740ab236e64503fba51be1101df7a1be83bf455d	synthetic_text_to_sql_test.snappy.parquet	1842	test	https://huggingface.co/datasets/gretelai/synthetic_text_to_sql/resolve/740ab236e64503fba51be1101df7a1be83bf455d/synthetic_text_to_sql_test.snappy.parquet	apache-2.0	Gretel Synthetic Text-to-SQL dataset creators; source card and author citation preserved in the download manifest.	proposed	What is the total number of patients treated with medication and therapy in each country?	CREATE TABLE patients (patient_id INT, name VARCHAR(50), age INT, state VARCHAR(50), country VARCHAR(50)); CREATE TABLE therapy_sessions (session_id INT, patient_id INT, therapist_id INT, session_date DATE); CREATE TABLE medications (medication_id INT, patient_id INT, medication_name VARCHAR(50), prescription_date DATE...	database_business_operations	database_business_operations.database_query_analysis	heuristic_unreviewed	Mapped from source sql_task_type='analytics and reporting'; no individual SQL execution or semantic review.	en	none	SELECT countries.country, COUNT(DISTINCT patients.patient_id) FROM patients JOIN medications ON patients.patient_id = medications.patient_id JOIN therapy_sessions ON patients.patient_id = therapy_sessions.patient_id JOIN (SELECT DISTINCT country FROM patients) AS countries ON patients.country = countries.country GROUP ...	{"domain":"mental health","domain_description":"In-depth data on mental health conditions, treatment approaches, patient outcomes, and public awareness campaigns in mental health.","id":1842,"sql_complexity":"multiple_joins","sql_complexity_description":"two or more joins (specify inner, outer, cross)","sql_explanation...	b74910cd756c88c45a4b51c1009905f96b7a77ba617a722b190ee8c1a134abe2	gretel_sql:test:1842	multiple_joins	intermediate	Estimated from source sql_complexity='multiple_joins'; query feature labels are only proxies for difficulty, not measured solver performance. No empirical solver calibration was performed.	heuristic	needs_difficulty_calibration	not_measured	analysis_only_unvalidated	primary	task_annotation	1,841	[ "difficulty_not_calibrated", "not_individually_reviewed", "reference_not_independently_verified", "reference_or_constraint_only_no_rubric", "response_diversity_not_validated", "source_evaluation_split", "sql_dialect_and_result_equivalence_need_review", "sql_not_executed", "synthetic_reference" ]	false	true	false	task:1ba7172f82a48399b6904ad5d93cd163e76720bafa1ebd6f19d7ff5d2fd6ad31	1	1	[ "gretel_sql" ]	held_out	curated_candidate	[]	core_candidate	[ "source_sql_complexity:multiple_joins", "sql_structure:selects=2;joins=3;statements=1;advanced=none", "combined_relational_operations_require_calibration" ]	structural_heuristic	Qwen3.6-27B	heuristic_not_model_measured
gretel_sql:5c45fd40432c81f39099bdf3fb85cd1cafd02b944ce515e1ee467979614321a6	task:5cc1f3056e37d477dc2994daff3e5d91dd654ec7f75923c04030655692d0126d	6a31f168d97ec790dc5ae2a5803ab6bd7f5c73ff19ef35c14534e856e59ce05f	gretel_sql	gretelai/synthetic_text_to_sql	740ab236e64503fba51be1101df7a1be83bf455d	synthetic_text_to_sql_test.snappy.parquet	1875	test	https://huggingface.co/datasets/gretelai/synthetic_text_to_sql/resolve/740ab236e64503fba51be1101df7a1be83bf455d/synthetic_text_to_sql_test.snappy.parquet	apache-2.0	Gretel Synthetic Text-to-SQL dataset creators; source card and author citation preserved in the download manifest.	proposed	Which incident response policies were implemented in Asia, and what is the rank of each policy based on the number of incidents it covers?	CREATE TABLE incidents (id INT, date DATE, category VARCHAR(20), source_ip VARCHAR(15), target_ip VARCHAR(15)); CREATE TABLE policies (id INT, date DATE, type VARCHAR(20), region VARCHAR(30)); INSERT INTO incidents (id, date, category, source_ip, target_ip) VALUES (1, '2021-01-01', 'malware', '192.168.1.100', '8.8.8.8'...	database_business_operations	database_business_operations.database_query_analysis	heuristic_unreviewed	Mapped from source sql_task_type='analytics and reporting'; no individual SQL execution or semantic review.	en	none	SELECT policies.type, RANK() OVER (ORDER BY incident_count DESC) as policy_rank FROM (SELECT policy.type, COUNT(*) as incident_count FROM incidents JOIN policies ON incidents.date = policies.date WHERE policies.region = 'Asia' GROUP BY policy.type) as subquery JOIN policies ON policies.type = subquery.type;	{"domain":"cybersecurity","domain_description":"Threat intelligence data, vulnerability assessments, security incident response metrics, and cybersecurity policy analysis.","id":1875,"sql_complexity":"window functions","sql_complexity_description":"window functions (e.g., ROW_NUMBER, LEAD, LAG, RANk, NTILE, PERCENT_RAN...	a35bd6e16975723fa4b6e058c335f0e9366e23ca3fcebacafa0e39fa853176cf	gretel_sql:test:1875	window functions	intermediate	Estimated from source sql_complexity='window functions'; query feature labels are only proxies for difficulty, not measured solver performance. No empirical solver calibration was performed.	heuristic	needs_difficulty_calibration	not_measured	analysis_only_unvalidated	primary	task_annotation	1,874	[ "difficulty_not_calibrated", "not_individually_reviewed", "reference_not_independently_verified", "reference_or_constraint_only_no_rubric", "response_diversity_not_validated", "source_evaluation_split", "sql_dialect_and_result_equivalence_need_review", "sql_not_executed", "synthetic_reference" ]	false	true	false	task:5cc1f3056e37d477dc2994daff3e5d91dd654ec7f75923c04030655692d0126d	1	1	[ "gretel_sql" ]	held_out	curated_candidate	[]	core_candidate	[ "source_sql_complexity:window functions", "sql_structure:selects=2;joins=2;statements=1;advanced=OVER", "combined_relational_operations_require_calibration" ]	structural_heuristic	Qwen3.6-27B	heuristic_not_model_measured
gretel_sql:9fa7767efb3637eb2369d4955b50334592c6237f50f05fc717d1f3f1a698433d	task:e848ac77c0fb6cfe41a6fc64e3a4147b69178421df181084dd587f953632567d	95128ba861fb5b0c6a1be7906e17e53903c91da4ba348db1c1435d3c7a3b0dc4	gretel_sql	gretelai/synthetic_text_to_sql	740ab236e64503fba51be1101df7a1be83bf455d	synthetic_text_to_sql_test.snappy.parquet	1876	test	https://huggingface.co/datasets/gretelai/synthetic_text_to_sql/resolve/740ab236e64503fba51be1101df7a1be83bf455d/synthetic_text_to_sql_test.snappy.parquet	apache-2.0	Gretel Synthetic Text-to-SQL dataset creators; source card and author citation preserved in the download manifest.	proposed	Which systems were scanned the most in the last month, and what were their maximum CVE scores?	CREATE TABLE systems (system_id INT, system_name VARCHAR(255));CREATE TABLE scan_dates (scan_date DATE, system_id INT);CREATE TABLE cve_scores (system_id INT, score INT, scan_date DATE);	database_business_operations	database_business_operations.database_query_analysis	heuristic_unreviewed	Mapped from source sql_task_type='analytics and reporting'; no individual SQL execution or semantic review.	en	none	SELECT s.system_name, MAX(c.score) as max_score FROM systems s INNER JOIN (SELECT system_id, COUNT(*) as scan_count FROM scan_dates GROUP BY system_id) sd ON s.system_id = sd.system_id LEFT JOIN cve_scores c ON s.system_id = c.system_id AND sd.scan_date = c.scan_date WHERE sd.scan_date >= DATE_SUB(CURRENT_DATE, INTERVA...	{"domain":"cybersecurity","domain_description":"Threat intelligence data, vulnerability assessments, security incident response metrics, and cybersecurity policy analysis.","id":1876,"sql_complexity":"multiple_joins","sql_complexity_description":"two or more joins (specify inner, outer, cross)","sql_explanation":"This ...	1188cd906b451a615ec5d9b4ba7047d9acee0c0803bcb5736972020e3c00b62a	gretel_sql:test:1876	multiple_joins	intermediate	Estimated from source sql_complexity='multiple_joins'; query feature labels are only proxies for difficulty, not measured solver performance. No empirical solver calibration was performed.	heuristic	needs_difficulty_calibration	not_measured	analysis_only_unvalidated	primary	task_annotation	1,875	[ "difficulty_not_calibrated", "not_individually_reviewed", "reference_not_independently_verified", "reference_or_constraint_only_no_rubric", "response_diversity_not_validated", "source_evaluation_split", "sql_dialect_and_result_equivalence_need_review", "sql_not_executed", "synthetic_reference" ]	false	true	false	task:e848ac77c0fb6cfe41a6fc64e3a4147b69178421df181084dd587f953632567d	1	1	[ "gretel_sql" ]	held_out	curated_candidate	[]	core_candidate	[ "source_sql_complexity:multiple_joins", "sql_structure:selects=2;joins=2;statements=1;advanced=none", "combined_relational_operations_require_calibration" ]	structural_heuristic	Qwen3.6-27B	heuristic_not_model_measured
gretel_sql:d5a6373ba18c69f68a07725b94f858ac7dc265988f8528eb36562e56998e4e1d	task:5ef3018a96bcc52a2762531f3586db49682c8b2cc8aac16ea849bda4621d5eee	091738afc07cfa1679e0222b0428c48bf8613fa55a0f802fab8eb76f3ad25b31	gretel_sql	gretelai/synthetic_text_to_sql	740ab236e64503fba51be1101df7a1be83bf455d	synthetic_text_to_sql_test.snappy.parquet	1914	test	https://huggingface.co/datasets/gretelai/synthetic_text_to_sql/resolve/740ab236e64503fba51be1101df7a1be83bf455d/synthetic_text_to_sql_test.snappy.parquet	apache-2.0	Gretel Synthetic Text-to-SQL dataset creators; source card and author citation preserved in the download manifest.	proposed	What is the average time between equipment maintenance for each type of military aircraft?	CREATE TABLE equipment (id INT, equipment_type VARCHAR(255), manufacturer VARCHAR(255)); CREATE TABLE maintenance (id INT, equipment_id INT, maintenance_date DATE); INSERT INTO equipment (id, equipment_type, manufacturer) VALUES (1, 'F-16', 'Lockheed Martin'); INSERT INTO equipment (id, equipment_type, manufacturer) VA...	database_business_operations	database_business_operations.database_query_analysis	heuristic_unreviewed	Mapped from source sql_task_type='analytics and reporting'; no individual SQL execution or semantic review.	en	none	SELECT e.equipment_type, AVG(DATEDIFF(m2.maintenance_date, m1.maintenance_date)) as avg_maintenance_interval FROM equipment e JOIN maintenance m1 ON e.id = m1.equipment_id JOIN maintenance m2 ON e.id = m2.equipment_id AND m2.maintenance_date > m1.maintenance_date GROUP BY e.equipment_type;	{"domain":"defense industry","domain_description":"Defense contract data, military equipment maintenance, threat intelligence metrics, and veteran employment stats.","id":1914,"sql_complexity":"multiple_joins","sql_complexity_description":"two or more joins (specify inner, outer, cross)","sql_explanation":"This SQL que...	445c5789f9b47f8ae76c511f28ed013c89c3c40f93609681325a342da5e89304	gretel_sql:test:1914	multiple_joins	intermediate	Estimated from source sql_complexity='multiple_joins'; query feature labels are only proxies for difficulty, not measured solver performance. No empirical solver calibration was performed.	heuristic	needs_difficulty_calibration	not_measured	analysis_only_unvalidated	primary	task_annotation	1,913	[ "difficulty_not_calibrated", "not_individually_reviewed", "reference_not_independently_verified", "reference_or_constraint_only_no_rubric", "response_diversity_not_validated", "source_evaluation_split", "sql_dialect_and_result_equivalence_need_review", "sql_not_executed", "synthetic_reference" ]	false	true	false	task:5ef3018a96bcc52a2762531f3586db49682c8b2cc8aac16ea849bda4621d5eee	1	1	[ "gretel_sql" ]	held_out	curated_candidate	[]	core_candidate	[ "source_sql_complexity:multiple_joins", "sql_structure:selects=1;joins=2;statements=1;advanced=none", "combined_relational_operations_require_calibration" ]	structural_heuristic	Qwen3.6-27B	heuristic_not_model_measured
gretel_sql:acc790cbac5b36a3f8120e9c3923f4a468693fbbb182fce10252720553785229	task:6744076195c0a70b8a9f57050f8c01feb7f729bbec378123dc85807abb3495dc	2f68e19c68d4cf9d687f871d39c4b79544310ac7a6a896b8b2ae6cc418e49e9b	gretel_sql	gretelai/synthetic_text_to_sql	740ab236e64503fba51be1101df7a1be83bf455d	synthetic_text_to_sql_test.snappy.parquet	1928	test	https://huggingface.co/datasets/gretelai/synthetic_text_to_sql/resolve/740ab236e64503fba51be1101df7a1be83bf455d/synthetic_text_to_sql_test.snappy.parquet	apache-2.0	Gretel Synthetic Text-to-SQL dataset creators; source card and author citation preserved in the download manifest.	proposed	Who were the top 3 volunteer programs by total hours in 2023?	CREATE TABLE VolunteerPrograms (ProgramID int, ProgramName varchar(255), VolunteerHours int); INSERT INTO VolunteerPrograms VALUES (1, 'Education', 1500), (2, 'Healthcare', 2000), (3, 'Environment', 1200), (4, 'Arts & Culture', 1750), (5, 'Social Services', 2200);	database_business_operations	database_business_operations.database_query_analysis	heuristic_unreviewed	Mapped from source sql_task_type='analytics and reporting'; no individual SQL execution or semantic review.	en	none	SELECT ProgramName FROM (SELECT ProgramName, ROW_NUMBER() OVER (ORDER BY VolunteerHours DESC) as Rank FROM VolunteerPrograms) as ProgramRanks WHERE Rank <= 3;	{"domain":"nonprofit operations","domain_description":"Donation records, program outcomes, volunteer engagement metrics, budget reports, and community impact assessments.","id":1928,"sql_complexity":"window functions","sql_complexity_description":"window functions (e.g., ROW_NUMBER, LEAD, LAG, RANk, NTILE, PERCENT_RANK...	1c07b8694c5c2881cf7c0c8453e98b001c491f01e67db7a1dcba87f70e889808	gretel_sql:test:1928	window functions	intermediate	Estimated from source sql_complexity='window functions'; query feature labels are only proxies for difficulty, not measured solver performance. No empirical solver calibration was performed.	heuristic	needs_difficulty_calibration	not_measured	analysis_only_unvalidated	primary	task_annotation	1,927	[ "difficulty_not_calibrated", "not_individually_reviewed", "reference_not_independently_verified", "reference_or_constraint_only_no_rubric", "response_diversity_not_validated", "source_evaluation_split", "sql_dialect_and_result_equivalence_need_review", "sql_not_executed", "synthetic_reference" ]	false	true	false	task:6744076195c0a70b8a9f57050f8c01feb7f729bbec378123dc85807abb3495dc	1	1	[ "gretel_sql" ]	held_out	curated_candidate	[]	core_candidate	[ "source_sql_complexity:window functions", "sql_structure:selects=2;joins=0;statements=1;advanced=OVER", "combined_relational_operations_require_calibration" ]	structural_heuristic	Qwen3.6-27B	heuristic_not_model_measured
gretel_sql:8cdc960c49afa303bfb22647d523a9caf09a8379cd80af8f4a49067dddf2afa7	task:2f6a6515031f6d5c57496b24618024c03ca3ea98cc01e053197d3e10c51df84d	38e22f30b5aeba353caecc87f24c5ba0e4136dda2af46d47efce03a8c5dae567	gretel_sql	gretelai/synthetic_text_to_sql	740ab236e64503fba51be1101df7a1be83bf455d	synthetic_text_to_sql_test.snappy.parquet	2048	test	https://huggingface.co/datasets/gretelai/synthetic_text_to_sql/resolve/740ab236e64503fba51be1101df7a1be83bf455d/synthetic_text_to_sql_test.snappy.parquet	apache-2.0	Gretel Synthetic Text-to-SQL dataset creators; source card and author citation preserved in the download manifest.	proposed	What is the average age of clinical psychologists who have treated mental health patients in Asia, ordered by the number of patients treated?	CREATE TABLE psychologists (id INT, name TEXT, age INT, country TEXT, patients INT); INSERT INTO psychologists (id, name, age, country, patients) VALUES (1, 'John Lee', 45, 'China', 75), (2, 'Mei Chen', 50, 'Japan', 60), (3, 'Kim Park', 40, 'South Korea', 80), (4, 'Anand Patel', 55, 'India', 90);	database_business_operations	database_business_operations.database_query_analysis	heuristic_unreviewed	Mapped from source sql_task_type='analytics and reporting'; no individual SQL execution or semantic review.	en	none	SELECT AVG(age) as avg_age FROM (SELECT age, ROW_NUMBER() OVER (PARTITION BY country ORDER BY patients DESC) as rn FROM psychologists WHERE country IN ('China', 'Japan', 'South Korea', 'India')) t WHERE rn = 1;	{"domain":"mental health","domain_description":"In-depth data on mental health conditions, treatment approaches, patient outcomes, and public awareness campaigns in mental health.","id":2048,"sql_complexity":"window functions","sql_complexity_description":"window functions (e.g., ROW_NUMBER, LEAD, LAG, RANk, NTILE, PER...	4c099e75e667d8fe85b9e69a17e105bc0e881edeca58cc578ee2bb7ea661e0e4	gretel_sql:test:2048	window functions	intermediate	Estimated from source sql_complexity='window functions'; query feature labels are only proxies for difficulty, not measured solver performance. No empirical solver calibration was performed.	heuristic	needs_difficulty_calibration	not_measured	analysis_only_unvalidated	primary	task_annotation	2,047	[ "difficulty_not_calibrated", "not_individually_reviewed", "reference_not_independently_verified", "reference_or_constraint_only_no_rubric", "response_diversity_not_validated", "source_evaluation_split", "sql_dialect_and_result_equivalence_need_review", "sql_not_executed", "synthetic_reference" ]	false	true	false	task:2f6a6515031f6d5c57496b24618024c03ca3ea98cc01e053197d3e10c51df84d	1	1	[ "gretel_sql" ]	held_out	curated_candidate	[]	core_candidate	[ "source_sql_complexity:window functions", "sql_structure:selects=2;joins=0;statements=1;advanced=OVER", "combined_relational_operations_require_calibration" ]	structural_heuristic	Qwen3.6-27B	heuristic_not_model_measured
gretel_sql:7035e8bee27c3dd152c5f3568e86b9fe5ca1b58a6c5132a47a9c75634bb7828c	task:3a43779c350fc766cdab586bcc225ce12eecdd8b2450481a621416555c07f683	8f91c7a8ae1ea1b74b7a950c4f2f299f84a3c85b45ecfe630ba26a2b5f4f73f0	gretel_sql	gretelai/synthetic_text_to_sql	740ab236e64503fba51be1101df7a1be83bf455d	synthetic_text_to_sql_test.snappy.parquet	2058	test	https://huggingface.co/datasets/gretelai/synthetic_text_to_sql/resolve/740ab236e64503fba51be1101df7a1be83bf455d/synthetic_text_to_sql_test.snappy.parquet	apache-2.0	Gretel Synthetic Text-to-SQL dataset creators; source card and author citation preserved in the download manifest.	proposed	Identify the number of mobile and broadband subscribers per region, and their respective percentage contributions to total subscribers in each service category.	CREATE TABLE MobileSubscribers (SubscriberID int, Region varchar(10), Service varchar(10)); CREATE TABLE BroadbandSubscribers (SubscriberID int, Region varchar(10), Service varchar(10)); INSERT INTO MobileSubscribers (SubscriberID, Region, Service) VALUES (1, 'North', 'mobile'), (2, 'North', 'mobile'), (3, 'South', 'mo...	database_business_operations	database_business_operations.database_query_analysis	heuristic_unreviewed	Mapped from source sql_task_type='analytics and reporting'; no individual SQL execution or semantic review.	en	none	SELECT R.Region, S.Service, COUNT(M.SubscriberID) AS MobileCount, COUNT(B.SubscriberID) AS BroadbandCount, (COUNT(M.SubscriberID)::float / (COUNT(M.SubscriberID) + COUNT(B.SubscriberID))) * 100 AS MobilePercent, (COUNT(B.SubscriberID)::float / (COUNT(M.SubscriberID) + COUNT(B.SubscriberID))) * 100 AS BroadbandPercent F...	{"domain":"telecommunications","domain_description":"Mobile and broadband subscriber data, network infrastructure investments, customer usage patterns, and regulatory compliance information.","id":2058,"sql_complexity":"multiple_joins","sql_complexity_description":"two or more joins (specify inner, outer, cross)","sql_...	03d862c7d7f9bc7c1a7e1ee9ee0c768c6dbffd037d3fa791ecd2440f48ad988a	gretel_sql:test:2058	multiple_joins	intermediate	Estimated from source sql_complexity='multiple_joins'; query feature labels are only proxies for difficulty, not measured solver performance. No empirical solver calibration was performed.	heuristic	needs_difficulty_calibration	not_measured	analysis_only_unvalidated	primary	task_annotation	2,057	[ "difficulty_not_calibrated", "not_individually_reviewed", "reference_not_independently_verified", "reference_or_constraint_only_no_rubric", "response_diversity_not_validated", "source_evaluation_split", "sql_dialect_and_result_equivalence_need_review", "sql_not_executed", "synthetic_reference" ]	false	true	false	task:3a43779c350fc766cdab586bcc225ce12eecdd8b2450481a621416555c07f683	1	1	[ "gretel_sql" ]	held_out	curated_candidate	[]	core_candidate	[ "source_sql_complexity:multiple_joins", "sql_structure:selects=1;joins=3;statements=1;advanced=none", "combined_relational_operations_require_calibration" ]	structural_heuristic	Qwen3.6-27B	heuristic_not_model_measured
gretel_sql:7abe1f8927c629524df2d05b110f0f14d250b8d0a62447eafdb07c54b030d3ef	task:491cd97876013cbce8561d1990859912bfb818d905f33035b3804d43a2f5add7	f3284e26512a15e62ebdd90c6e898a58d740357fbcd68022db163a9549732270	gretel_sql	gretelai/synthetic_text_to_sql	740ab236e64503fba51be1101df7a1be83bf455d	synthetic_text_to_sql_test.snappy.parquet	2089	test	https://huggingface.co/datasets/gretelai/synthetic_text_to_sql/resolve/740ab236e64503fba51be1101df7a1be83bf455d/synthetic_text_to_sql_test.snappy.parquet	apache-2.0	Gretel Synthetic Text-to-SQL dataset creators; source card and author citation preserved in the download manifest.	proposed	Find the number of workers and projects for sustainable building permits	CREATE TABLE project (id INT PRIMARY KEY, name VARCHAR(50), location VARCHAR(50), start_date DATE); INSERT INTO project (id, name, location, start_date) VALUES (1, 'GreenTowers', 'NYC', '2020-01-01'); CREATE TABLE building_permit (id INT PRIMARY KEY, project_id INT, permit_type VARCHAR(50), permit_number INT);	database_business_operations	database_business_operations.database_query_analysis	heuristic_unreviewed	Mapped from source sql_task_type='analytics and reporting'; no individual SQL execution or semantic review.	en	none	SELECT COUNT(DISTINCT p.id) AS projects_count, SUM(l.workers_count) AS total_workers_count FROM project p INNER JOIN labor_statistics l ON p.id = l.project_id INNER JOIN building_permit bp ON p.id = bp.project_id WHERE bp.permit_type = 'Sustainable';	{"domain":"construction","domain_description":"Building permit data, construction labor statistics, project timeline information, and sustainable building practices.","id":2089,"sql_complexity":"multiple_joins","sql_complexity_description":"two or more joins (specify inner, outer, cross)","sql_explanation":"This query ...	6f23348f4c98f3eee4464dca780439d2fc6d064d43e523a8162ef4d871c8391d	gretel_sql:test:2089	multiple_joins	intermediate	Estimated from source sql_complexity='multiple_joins'; query feature labels are only proxies for difficulty, not measured solver performance. No empirical solver calibration was performed.	heuristic	needs_difficulty_calibration	not_measured	analysis_only_unvalidated	primary	task_annotation	2,088	[ "difficulty_not_calibrated", "not_individually_reviewed", "reference_not_independently_verified", "reference_or_constraint_only_no_rubric", "response_diversity_not_validated", "source_evaluation_split", "sql_dialect_and_result_equivalence_need_review", "sql_not_executed", "synthetic_reference" ]	false	true	false	task:491cd97876013cbce8561d1990859912bfb818d905f33035b3804d43a2f5add7	1	1	[ "gretel_sql" ]	held_out	curated_candidate	[]	core_candidate	[ "source_sql_complexity:multiple_joins", "sql_structure:selects=1;joins=2;statements=1;advanced=none", "combined_relational_operations_require_calibration" ]	structural_heuristic	Qwen3.6-27B	heuristic_not_model_measured
gretel_sql:0951363747a82c075e4ecebafc92cc44d049578485761c7984b57bc0551781e7	task:15250d7b8169414a9f72f836e29d0aa2eab7827f58accbe8310b58195357fc91	cf4faad5cc7bcfc51a63e3baeaa381a485866ac6b5ce927ff062276ec13b8968	gretel_sql	gretelai/synthetic_text_to_sql	740ab236e64503fba51be1101df7a1be83bf455d	synthetic_text_to_sql_test.snappy.parquet	2123	test	https://huggingface.co/datasets/gretelai/synthetic_text_to_sql/resolve/740ab236e64503fba51be1101df7a1be83bf455d/synthetic_text_to_sql_test.snappy.parquet	apache-2.0	Gretel Synthetic Text-to-SQL dataset creators; source card and author citation preserved in the download manifest.	proposed	What is the total patient count for each mental health condition, for providers in the Northeast and Southeast?	CREATE TABLE mental_health_conditions (condition_id INT, condition_name VARCHAR(50)); INSERT INTO mental_health_conditions (condition_id, condition_name) VALUES (1, 'Anxiety'), (2, 'Depression'), (3, 'Bipolar Disorder'); CREATE TABLE providers (provider_id INT, provider_name VARCHAR(50), region_id INT); INSERT INTO pro...	database_business_operations	database_business_operations.database_query_analysis	heuristic_unreviewed	Mapped from source sql_task_type='analytics and reporting'; no individual SQL execution or semantic review.	en	none	SELECT mhc.condition_name, SUM(pp.patient_id) as total_patients FROM mental_health_conditions mhc JOIN provider_patients pp ON mhc.condition_id = pp.condition_id JOIN providers p ON pp.provider_id = p.provider_id WHERE p.region_id IN (1, 2) GROUP BY mhc.condition_name;	{"domain":"healthcare","domain_description":"Healthcare data on mental health parity, cultural competency, health equity metrics, and community health workers.","id":2123,"sql_complexity":"multiple_joins","sql_complexity_description":"two or more joins (specify inner, outer, cross)","sql_explanation":"This query perfor...	786af8b92987add7f0395f9dbe177ac8fc74f70d12aeea62bf901e6fbe71259a	gretel_sql:test:2123	multiple_joins	intermediate	Estimated from source sql_complexity='multiple_joins'; query feature labels are only proxies for difficulty, not measured solver performance. No empirical solver calibration was performed.	heuristic	needs_difficulty_calibration	not_measured	analysis_only_unvalidated	primary	task_annotation	2,122	[ "difficulty_not_calibrated", "not_individually_reviewed", "reference_not_independently_verified", "reference_or_constraint_only_no_rubric", "response_diversity_not_validated", "source_evaluation_split", "sql_dialect_and_result_equivalence_need_review", "sql_not_executed", "synthetic_reference" ]	false	true	false	task:15250d7b8169414a9f72f836e29d0aa2eab7827f58accbe8310b58195357fc91	1	1	[ "gretel_sql" ]	held_out	curated_candidate	[]	core_candidate	[ "source_sql_complexity:multiple_joins", "sql_structure:selects=1;joins=2;statements=1;advanced=none", "combined_relational_operations_require_calibration" ]	structural_heuristic	Qwen3.6-27B	heuristic_not_model_measured
gretel_sql:3dcf906ef42a2320fc148a1d4a3b7705dcd6afcf105fbb4b2501b85f05b8b091	task:53d8488a62b8f9f1a59d7f24f13cca58c54fb146c67f5a031b641cf5ecfa2027	997618c24a1a5ef34e341b066effb7e196df203521b5b3c88973a9e09a467b1e	gretel_sql	gretelai/synthetic_text_to_sql	740ab236e64503fba51be1101df7a1be83bf455d	synthetic_text_to_sql_test.snappy.parquet	2128	test	https://huggingface.co/datasets/gretelai/synthetic_text_to_sql/resolve/740ab236e64503fba51be1101df7a1be83bf455d/synthetic_text_to_sql_test.snappy.parquet	apache-2.0	Gretel Synthetic Text-to-SQL dataset creators; source card and author citation preserved in the download manifest.	proposed	What is the success rate of cognitive behavioral therapy (CBT) for patients with depression in the African American community?	CREATE TABLE therapy_approaches (approach_id INT, name VARCHAR(255)); CREATE TABLE patients (patient_id INT, age INT, gender VARCHAR(10), condition VARCHAR(255), ethnicity VARCHAR(255)); CREATE TABLE therapy_sessions (session_id INT, patient_id INT, therapist_id INT, session_date DATE, success BOOLEAN, approach_id INT)...	database_business_operations	database_business_operations.database_query_analysis	heuristic_unreviewed	Mapped from source sql_task_type='analytics and reporting'; no individual SQL execution or semantic review.	en	none	SELECT AVG(CASE WHEN therapy_sessions.success THEN 1 ELSE 0 END) AS success_rate FROM therapy_sessions JOIN patients ON therapy_sessions.patient_id = patients.patient_id JOIN therapy_approaches ON therapy_sessions.approach_id = therapy_approaches.approach_id WHERE patients.condition = 'depression' AND patients.ethnicit...	{"domain":"mental health","domain_description":"In-depth data on mental health conditions, treatment approaches, patient outcomes, and public awareness campaigns in mental health.","id":2128,"sql_complexity":"multiple_joins","sql_complexity_description":"two or more joins (specify inner, outer, cross)","sql_explanation...	bfb551572c1793bbd43164b9864e771bded03d05ea3d75ed34d90497f93335ee	gretel_sql:test:2128	multiple_joins	intermediate	Estimated from source sql_complexity='multiple_joins'; query feature labels are only proxies for difficulty, not measured solver performance. No empirical solver calibration was performed.	heuristic	needs_difficulty_calibration	not_measured	analysis_only_unvalidated	primary	task_annotation	2,127	[ "difficulty_not_calibrated", "not_individually_reviewed", "reference_not_independently_verified", "reference_or_constraint_only_no_rubric", "response_diversity_not_validated", "source_evaluation_split", "sql_dialect_and_result_equivalence_need_review", "sql_not_executed", "synthetic_reference" ]	false	true	false	task:53d8488a62b8f9f1a59d7f24f13cca58c54fb146c67f5a031b641cf5ecfa2027	1	1	[ "gretel_sql" ]	held_out	curated_candidate	[]	core_candidate	[ "source_sql_complexity:multiple_joins", "sql_structure:selects=1;joins=2;statements=1;advanced=none", "combined_relational_operations_require_calibration" ]	structural_heuristic	Qwen3.6-27B	heuristic_not_model_measured
gretel_sql:bfe186cd782de6177d2674aedb28634ceaf75332ccee73d8668489a2c59a3324	task:d487ecbafcfd151d5ae8a91b32a2fb9b8dfe4bb0952de88cebbecbe0e61b674d	73b2586f9b1f56cab4f4026d5afa66f845951720ba36206acf62e3f571c0ada1	gretel_sql	gretelai/synthetic_text_to_sql	740ab236e64503fba51be1101df7a1be83bf455d	synthetic_text_to_sql_test.snappy.parquet	2129	test	https://huggingface.co/datasets/gretelai/synthetic_text_to_sql/resolve/740ab236e64503fba51be1101df7a1be83bf455d/synthetic_text_to_sql_test.snappy.parquet	apache-2.0	Gretel Synthetic Text-to-SQL dataset creators; source card and author citation preserved in the download manifest.	proposed	Which mental health conditions have the highest success rates for treatment, and how many patients have been treated for each?	CREATE TABLE mental_health_conditions (id INT, name VARCHAR(50), prevalence FLOAT); CREATE TABLE treatments (id INT, condition_id INT, name VARCHAR(50), approach VARCHAR(50), success_rate FLOAT); CREATE TABLE patient_outcomes (id INT, treatment_id INT, patient_id INT);	database_business_operations	database_business_operations.database_query_analysis	heuristic_unreviewed	Mapped from source sql_task_type='analytics and reporting'; no individual SQL execution or semantic review.	en	none	SELECT mhc.name, t.name, COUNT(po.patient_id) as patient_count FROM mental_health_conditions mhc JOIN treatments t ON mhc.id = t.condition_id JOIN patient_outcomes po ON t.id = po.treatment_id GROUP BY mhc.name, t.name ORDER BY t.success_rate DESC;	{"domain":"mental health","domain_description":"In-depth data on mental health conditions, treatment approaches, patient outcomes, and public awareness campaigns in mental health.","id":2129,"sql_complexity":"multiple_joins","sql_complexity_description":"two or more joins (specify inner, outer, cross)","sql_explanation...	1f071238188332b34a8be512ba94a2fa8da13198a76d548e954354da50d4571e	gretel_sql:test:2129	multiple_joins	intermediate	Estimated from source sql_complexity='multiple_joins'; query feature labels are only proxies for difficulty, not measured solver performance. No empirical solver calibration was performed.	heuristic	needs_difficulty_calibration	not_measured	analysis_only_unvalidated	primary	task_annotation	2,128	[ "difficulty_not_calibrated", "not_individually_reviewed", "reference_not_independently_verified", "reference_or_constraint_only_no_rubric", "response_diversity_not_validated", "source_evaluation_split", "sql_dialect_and_result_equivalence_need_review", "sql_not_executed", "synthetic_reference" ]	false	true	false	task:d487ecbafcfd151d5ae8a91b32a2fb9b8dfe4bb0952de88cebbecbe0e61b674d	1	1	[ "gretel_sql" ]	held_out	curated_candidate	[]	core_candidate	[ "source_sql_complexity:multiple_joins", "sql_structure:selects=1;joins=2;statements=1;advanced=none", "combined_relational_operations_require_calibration" ]	structural_heuristic	Qwen3.6-27B	heuristic_not_model_measured
gretel_sql:82ec2c9cff05cacf2097e20a164c25211376e8f8820485f68df2889231a83944	task:7b241a0cea5bb620dc8623fc999bbcff1a072ed1dcf2f0d4d58e34128396800b	8ee9dabd2f384f345a086d5beaed663c40e75b17f47ae25d3c5e927454ae16ce	gretel_sql	gretelai/synthetic_text_to_sql	740ab236e64503fba51be1101df7a1be83bf455d	synthetic_text_to_sql_test.snappy.parquet	2143	test	https://huggingface.co/datasets/gretelai/synthetic_text_to_sql/resolve/740ab236e64503fba51be1101df7a1be83bf455d/synthetic_text_to_sql_test.snappy.parquet	apache-2.0	Gretel Synthetic Text-to-SQL dataset creators; source card and author citation preserved in the download manifest.	proposed	What is the total revenue generated from postpaid mobile plans in the Midwest region for the year 2022?	CREATE TABLE subscribers(id INT, plan_type VARCHAR(10), region VARCHAR(10)); INSERT INTO subscribers VALUES (1, 'postpaid', 'Midwest'); CREATE TABLE plans(plan_type VARCHAR(10), price DECIMAL(5,2)); INSERT INTO plans VALUES ('postpaid', 50.00); CREATE TABLE transactions(subscriber_id INT, transaction_date DATE, plan_id...	database_business_operations	database_business_operations.database_query_analysis	heuristic_unreviewed	Mapped from source sql_task_type='analytics and reporting'; no individual SQL execution or semantic review.	en	none	SELECT SUM(plans.price) FROM subscribers INNER JOIN transactions ON subscribers.id = transactions.subscriber_id INNER JOIN plans ON subscribers.plan_type = plans.plan_type WHERE subscribers.region = 'Midwest' AND YEAR(transactions.transaction_date) = 2022 AND subscribers.plan_type = 'postpaid';	{"domain":"telecommunications","domain_description":"Mobile and broadband subscriber data, network infrastructure investments, customer usage patterns, and regulatory compliance information.","id":2143,"sql_complexity":"multiple_joins","sql_complexity_description":"two or more joins (specify inner, outer, cross)","sql_...	1c66857ce10bc25745220b6f9b67f0d6921f19baeb776c8e2b2d1026528fc941	gretel_sql:test:2143	multiple_joins	intermediate	Estimated from source sql_complexity='multiple_joins'; query feature labels are only proxies for difficulty, not measured solver performance. No empirical solver calibration was performed.	heuristic	needs_difficulty_calibration	not_measured	analysis_only_unvalidated	primary	task_annotation	2,142	[ "difficulty_not_calibrated", "not_individually_reviewed", "reference_not_independently_verified", "reference_or_constraint_only_no_rubric", "response_diversity_not_validated", "source_evaluation_split", "sql_dialect_and_result_equivalence_need_review", "sql_not_executed", "synthetic_reference" ]	false	true	false	task:7b241a0cea5bb620dc8623fc999bbcff1a072ed1dcf2f0d4d58e34128396800b	1	1	[ "gretel_sql" ]	held_out	curated_candidate	[]	core_candidate	[ "source_sql_complexity:multiple_joins", "sql_structure:selects=1;joins=2;statements=1;advanced=none", "combined_relational_operations_require_calibration" ]	structural_heuristic	Qwen3.6-27B	heuristic_not_model_measured
gretel_sql:36e5a1c14c5100ddced7a777b41589bf14e253e5402c5231688dcb38615cc646	task:cbbf1be5e7879edde22adcce10bfafddfcf9fbe72a567fbd4942ffcc51b500b3	f2e404d3bd9873a8e294e4749e26f35e8983a05feb4e14f04f251dab8085e914	gretel_sql	gretelai/synthetic_text_to_sql	740ab236e64503fba51be1101df7a1be83bf455d	synthetic_text_to_sql_test.snappy.parquet	2158	test	https://huggingface.co/datasets/gretelai/synthetic_text_to_sql/resolve/740ab236e64503fba51be1101df7a1be83bf455d/synthetic_text_to_sql_test.snappy.parquet	apache-2.0	Gretel Synthetic Text-to-SQL dataset creators; source card and author citation preserved in the download manifest.	proposed	What is the total revenue generated by retail stores located in New York that sell sustainable products?	CREATE TABLE RetailStores (StoreID INT, StoreName VARCHAR(50), State VARCHAR(50)); INSERT INTO RetailStores (StoreID, StoreName, State) VALUES (1, 'RetailStoreA', 'New York'); CREATE TABLE Sales (SaleID INT, StoreID INT, ProductID INT, Quantity INT, Price DECIMAL(5,2)); INSERT INTO Sales (SaleID, StoreID, ProductID, Qu...	database_business_operations	database_business_operations.database_query_analysis	heuristic_unreviewed	Mapped from source sql_task_type='analytics and reporting'; no individual SQL execution or semantic review.	en	none	SELECT SUM(Quantity * Price) FROM Sales JOIN RetailStores ON Sales.StoreID = RetailStores.StoreID JOIN Products ON Sales.ProductID = Products.ProductID WHERE RetailStores.State = 'New York' AND Products.IsSustainable = true;	{"domain":"retail","domain_description":"Retail data on circular supply chains, ethical labor practices, product transparency, and consumer education.","id":2158,"sql_complexity":"multiple_joins","sql_complexity_description":"two or more joins (specify inner, outer, cross)","sql_explanation":"This query calculates the ...	6f040a4b41b693fd9d8247b0d7bb78dae5c6098685296d482691567ec60399b3	gretel_sql:test:2158	multiple_joins	intermediate	Estimated from source sql_complexity='multiple_joins'; query feature labels are only proxies for difficulty, not measured solver performance. No empirical solver calibration was performed.	heuristic	needs_difficulty_calibration	not_measured	analysis_only_unvalidated	primary	task_annotation	2,157	[ "difficulty_not_calibrated", "not_individually_reviewed", "reference_not_independently_verified", "reference_or_constraint_only_no_rubric", "response_diversity_not_validated", "source_evaluation_split", "sql_dialect_and_result_equivalence_need_review", "sql_not_executed", "synthetic_reference" ]	false	true	false	task:cbbf1be5e7879edde22adcce10bfafddfcf9fbe72a567fbd4942ffcc51b500b3	1	1	[ "gretel_sql" ]	held_out	curated_candidate	[]	core_candidate	[ "source_sql_complexity:multiple_joins", "sql_structure:selects=1;joins=2;statements=1;advanced=none", "combined_relational_operations_require_calibration" ]	structural_heuristic	Qwen3.6-27B	heuristic_not_model_measured
gretel_sql:d8b7f29415c60157f9f91fd93c7327735636b5c345790663e98132b52006280f	task:52a605959f41c0a9933a5b0158da293131061b9deffe136a322e966fc18a14ef	17446c7dfbedaddbb887beda4636a1c4d1e72d792369b9772a8bfb7c826a3d3f	gretel_sql	gretelai/synthetic_text_to_sql	740ab236e64503fba51be1101df7a1be83bf455d	synthetic_text_to_sql_test.snappy.parquet	2216	test	https://huggingface.co/datasets/gretelai/synthetic_text_to_sql/resolve/740ab236e64503fba51be1101df7a1be83bf455d/synthetic_text_to_sql_test.snappy.parquet	apache-2.0	Gretel Synthetic Text-to-SQL dataset creators; source card and author citation preserved in the download manifest.	proposed	What is the virtual tour engagement rate for the top 2 countries with the highest engagement rates, ordered by engagement rate in descending order?	CREATE TABLE virtual_tours (tour_id INT, hotel_name TEXT, country TEXT, engagement_rate FLOAT); INSERT INTO virtual_tours (tour_id, hotel_name, country, engagement_rate) VALUES (1, 'Hotel A', 'USA', 0.06), (2, 'Hotel B', 'Canada', 0.08), (3, 'Hotel C', 'Mexico', 0.05), (4, 'Hotel D', 'USA', 0.07);	database_business_operations	database_business_operations.database_query_analysis	heuristic_unreviewed	Mapped from source sql_task_type='analytics and reporting'; no individual SQL execution or semantic review.	en	none	SELECT country, engagement_rate FROM (SELECT country, engagement_rate, RANK() OVER (ORDER BY engagement_rate DESC) as rank FROM virtual_tours) as subquery WHERE rank <= 2 ORDER BY engagement_rate DESC;	{"domain":"hospitality technology","domain_description":"Hotel tech adoption metrics, online travel agency data, virtual tour engagement stats, and hospitality AI trends.","id":2216,"sql_complexity":"window functions","sql_complexity_description":"window functions (e.g., ROW_NUMBER, LEAD, LAG, RANk, NTILE, PERCENT_RANK...	25269c1126184f07fe35f1949c4121f353744a9f879991b2f0407104b0b91f2b	gretel_sql:test:2216	window functions	intermediate	Estimated from source sql_complexity='window functions'; query feature labels are only proxies for difficulty, not measured solver performance. No empirical solver calibration was performed.	heuristic	needs_difficulty_calibration	not_measured	analysis_only_unvalidated	primary	task_annotation	2,215	[ "difficulty_not_calibrated", "not_individually_reviewed", "reference_not_independently_verified", "reference_or_constraint_only_no_rubric", "response_diversity_not_validated", "source_evaluation_split", "sql_dialect_and_result_equivalence_need_review", "sql_not_executed", "synthetic_reference" ]	false	true	false	task:52a605959f41c0a9933a5b0158da293131061b9deffe136a322e966fc18a14ef	1	1	[ "gretel_sql" ]	held_out	curated_candidate	[]	core_candidate	[ "source_sql_complexity:window functions", "sql_structure:selects=2;joins=0;statements=1;advanced=OVER", "combined_relational_operations_require_calibration" ]	structural_heuristic	Qwen3.6-27B	heuristic_not_model_measured
gretel_sql:894546b13fd2bb6246910207fa3b9220eaeecbded1db445e129258d7b74ab755	task:c5a88f13ef1fb47a46a856de544c78a0b13288df81c0d7075f0439ab3cad338b	b8d1dd2ab2836e5a55f9a59838703c346011485c0ba87f25455d0fe5d285755e	gretel_sql	gretelai/synthetic_text_to_sql	740ab236e64503fba51be1101df7a1be83bf455d	synthetic_text_to_sql_test.snappy.parquet	2219	test	https://huggingface.co/datasets/gretelai/synthetic_text_to_sql/resolve/740ab236e64503fba51be1101df7a1be83bf455d/synthetic_text_to_sql_test.snappy.parquet	apache-2.0	Gretel Synthetic Text-to-SQL dataset creators; source card and author citation preserved in the download manifest.	proposed	What is the average population of cities with a UNESCO World Heritage designation, ordered by designation date?	CREATE TABLE cities (name VARCHAR(255), population INT, designation_date DATE); INSERT INTO cities (name, population, designation_date) VALUES ('Paris', 2141000, '1991-09-16'); INSERT INTO cities (name, population, designation_date) VALUES ('Rio de Janeiro', 6727000, '2012-07-01');	database_business_operations	database_business_operations.database_query_analysis	heuristic_unreviewed	Mapped from source sql_task_type='analytics and reporting'; no individual SQL execution or semantic review.	en	none	SELECT AVG(population) FROM (SELECT population, ROW_NUMBER() OVER (ORDER BY designation_date) rn FROM cities WHERE name IN (SELECT name FROM heritagesites)) t WHERE rn % 2 = 1;	{"domain":"cultural preservation","domain_description":"In-depth data on heritage sites, traditional arts, language preservation, and community engagement in cultural preservation.","id":2219,"sql_complexity":"window functions","sql_complexity_description":"window functions (e.g., ROW_NUMBER, LEAD, LAG, RANk, NTILE, PE...	e8725e75a24b577abaa489f9c92ce651b7d1751dad772b8a702e0e116639a6ba	gretel_sql:test:2219	window functions	intermediate	Estimated from source sql_complexity='window functions'; query feature labels are only proxies for difficulty, not measured solver performance. No empirical solver calibration was performed.	heuristic	needs_difficulty_calibration	not_measured	analysis_only_unvalidated	primary	task_annotation	2,218	[ "difficulty_not_calibrated", "not_individually_reviewed", "reference_not_independently_verified", "reference_or_constraint_only_no_rubric", "response_diversity_not_validated", "source_evaluation_split", "sql_dialect_and_result_equivalence_need_review", "sql_not_executed", "synthetic_reference" ]	false	true	false	task:c5a88f13ef1fb47a46a856de544c78a0b13288df81c0d7075f0439ab3cad338b	1	1	[ "gretel_sql" ]	held_out	curated_candidate	[]	core_candidate	[ "source_sql_complexity:window functions", "sql_structure:selects=3;joins=0;statements=1;advanced=OVER", "combined_relational_operations_require_calibration" ]	structural_heuristic	Qwen3.6-27B	heuristic_not_model_measured
gretel_sql:6fa36c82e82711d1d66682dac33baa0e9b2dcdbb9fd759f9554bcbcc6bab9502	task:7d0d5384cecd4d0cfa86863f7454be19b7ee6588ae757fce3aee34a27396ba86	626219aa0da2c71ff8689ee07529872514bdb9ef8e6bee30db8cec9290e56857	gretel_sql	gretelai/synthetic_text_to_sql	740ab236e64503fba51be1101df7a1be83bf455d	synthetic_text_to_sql_test.snappy.parquet	2261	test	https://huggingface.co/datasets/gretelai/synthetic_text_to_sql/resolve/740ab236e64503fba51be1101df7a1be83bf455d/synthetic_text_to_sql_test.snappy.parquet	apache-2.0	Gretel Synthetic Text-to-SQL dataset creators; source card and author citation preserved in the download manifest.	proposed	What is the total number of animals that have been released into 'protected' habitats, and the average weight of those animals?	CREATE TABLE habitats (habitat_id INT, habitat_name VARCHAR(50), habitat_status VARCHAR(50)); INSERT INTO habitats (habitat_id, habitat_name, habitat_status) VALUES (1, 'Habitat A', 'protected'), (2, 'Habitat B', 'unprotected'); CREATE TABLE animal_habitats (animal_id INT, habitat_id INT, animal_weight FLOAT); INSERT I...	database_business_operations	database_business_operations.database_query_analysis	heuristic_unreviewed	Mapped from source sql_task_type='analytics and reporting'; no individual SQL execution or semantic review.	en	none	SELECT COUNT(*), AVG(animal_habitats.animal_weight) FROM animal_habitats INNER JOIN animals ON animal_habitats.animal_id = animals.animal_id INNER JOIN habitats ON animal_habitats.habitat_id = habitats.habitat_id WHERE habitats.habitat_status = 'protected';	{"domain":"wildlife conservation","domain_description":"Animal population data, habitat preservation efforts, and community education programs.","id":2261,"sql_complexity":"multiple_joins","sql_complexity_description":"two or more joins (specify inner, outer, cross)","sql_explanation":"This query counts the total numbe...	0ba8959a8a8444d0816dbca3c88c61fb705be249b45bb86f1c6f3462ff1fb2a3	gretel_sql:test:2261	multiple_joins	intermediate	Estimated from source sql_complexity='multiple_joins'; query feature labels are only proxies for difficulty, not measured solver performance. No empirical solver calibration was performed.	heuristic	needs_difficulty_calibration	not_measured	analysis_only_unvalidated	primary	task_annotation	2,260	[ "difficulty_not_calibrated", "not_individually_reviewed", "reference_not_independently_verified", "reference_or_constraint_only_no_rubric", "response_diversity_not_validated", "source_evaluation_split", "sql_dialect_and_result_equivalence_need_review", "sql_not_executed", "synthetic_reference" ]	false	true	false	task:7d0d5384cecd4d0cfa86863f7454be19b7ee6588ae757fce3aee34a27396ba86	1	1	[ "gretel_sql" ]	held_out	curated_candidate	[]	core_candidate	[ "source_sql_complexity:multiple_joins", "sql_structure:selects=1;joins=2;statements=1;advanced=none", "combined_relational_operations_require_calibration" ]	structural_heuristic	Qwen3.6-27B	heuristic_not_model_measured
gretel_sql:af1aabadb25fd92c0c95d6e9f33dedafa8ade33a6dbf979ad2cb1b881afd426a	task:a84962f4c3b1234245fa4b3ba323a08929f2e71a116db3cc11371844cb9362d9	db0884e9360ebd0756e6f6fa1d278d40b91b834bf393c6af6d0bdb1d9b6dc0fe	gretel_sql	gretelai/synthetic_text_to_sql	740ab236e64503fba51be1101df7a1be83bf455d	synthetic_text_to_sql_test.snappy.parquet	2304	test	https://huggingface.co/datasets/gretelai/synthetic_text_to_sql/resolve/740ab236e64503fba51be1101df7a1be83bf455d/synthetic_text_to_sql_test.snappy.parquet	apache-2.0	Gretel Synthetic Text-to-SQL dataset creators; source card and author citation preserved in the download manifest.	proposed	Find the top 5 countries with the lowest financial capability scores in Europe.	CREATE TABLE financial_capability (id INT, country VARCHAR(255), score INT);	database_business_operations	database_business_operations.database_query_analysis	heuristic_unreviewed	Mapped from source sql_task_type='analytics and reporting'; no individual SQL execution or semantic review.	en	none	SELECT country, score FROM (SELECT country, score, ROW_NUMBER() OVER (ORDER BY score ASC) rn FROM financial_capability WHERE region = 'Europe') t WHERE rn <= 5;	{"domain":"finance","domain_description":"Financial data on Shariah-compliant finance, socially responsible lending, financial capability, and financial wellbeing.","id":2304,"sql_complexity":"window functions","sql_complexity_description":"window functions (e.g., ROW_NUMBER, LEAD, LAG, RANk, NTILE, PERCENT_RANK, etc.)...	b416b1a75ae15c7c4dd200f959bc3b5bd8a9be36f6d2ac8b22a0ff158615fa1d	gretel_sql:test:2304	window functions	intermediate	Estimated from source sql_complexity='window functions'; query feature labels are only proxies for difficulty, not measured solver performance. No empirical solver calibration was performed.	heuristic	needs_difficulty_calibration	not_measured	analysis_only_unvalidated	primary	task_annotation	2,303	[ "difficulty_not_calibrated", "not_individually_reviewed", "reference_not_independently_verified", "reference_or_constraint_only_no_rubric", "response_diversity_not_validated", "source_evaluation_split", "sql_dialect_and_result_equivalence_need_review", "sql_not_executed", "synthetic_reference" ]	false	true	false	task:a84962f4c3b1234245fa4b3ba323a08929f2e71a116db3cc11371844cb9362d9	1	1	[ "gretel_sql" ]	held_out	curated_candidate	[]	core_candidate	[ "source_sql_complexity:window functions", "sql_structure:selects=2;joins=0;statements=1;advanced=OVER", "combined_relational_operations_require_calibration" ]	structural_heuristic	Qwen3.6-27B	heuristic_not_model_measured
gretel_sql:07a13e82d4621ff567e760020b38305a4167d420cc0a321c08d4531969842fe6	task:39fa8f33a9305e900ca13675fb6bac0170cca8782973790a1ceb11768a8eaf9c	e975f12ad5d1b120ad35e2495ca69d8994281279d833bd3aa7279f9a2d2751c3	gretel_sql	gretelai/synthetic_text_to_sql	740ab236e64503fba51be1101df7a1be83bf455d	synthetic_text_to_sql_test.snappy.parquet	2340	test	https://huggingface.co/datasets/gretelai/synthetic_text_to_sql/resolve/740ab236e64503fba51be1101df7a1be83bf455d/synthetic_text_to_sql_test.snappy.parquet	apache-2.0	Gretel Synthetic Text-to-SQL dataset creators; source card and author citation preserved in the download manifest.	proposed	Show the names and negotiation dates for all unions in New York that had at least 3 machinery malfunctions in the last 6 months.	CREATE TABLE UnionMembers (id INT PRIMARY KEY, name VARCHAR(50), state VARCHAR(2), union_id INT, FOREIGN KEY (union_id) REFERENCES UnionNegotiations(union_id)); CREATE TABLE UnionNegotiations (id INT PRIMARY KEY, union_id INT, negotiation_date DATE); CREATE TABLE WorkplaceSafety (id INT PRIMARY KEY, union_id INT, incid...	database_business_operations	database_business_operations.database_query_analysis	heuristic_unreviewed	Mapped from source sql_task_type='analytics and reporting'; no individual SQL execution or semantic review.	en	none	SELECT m.name, n.negotiation_date FROM UnionMembers m JOIN UnionNegotiations n ON m.union_id = n.union_id JOIN (SELECT w.union_id FROM WorkplaceSafety w WHERE w.incident_date >= DATE(NOW()) - INTERVAL 6 MONTH AND w.incident_type = 'Machinery Malfunction' GROUP BY w.union_id HAVING COUNT(*) >= 3) malfunctions ON m.union...	{"domain":"trade unions","domain_description":"Union membership statistics, collective bargaining data, labor rights advocacy, and workplace safety metrics.","id":2340,"sql_complexity":"multiple_joins","sql_complexity_description":"two or more joins (specify inner, outer, cross)","sql_explanation":"This query uses a ne...	48316dab39bbb318ad541d19e863abbc165443a9fcc4c9eafc16ca06cdcb65da	gretel_sql:test:2340	multiple_joins	intermediate	Estimated from source sql_complexity='multiple_joins'; query feature labels are only proxies for difficulty, not measured solver performance. No empirical solver calibration was performed.	heuristic	needs_difficulty_calibration	not_measured	analysis_only_unvalidated	primary	task_annotation	2,339	[ "difficulty_not_calibrated", "not_individually_reviewed", "reference_not_independently_verified", "reference_or_constraint_only_no_rubric", "response_diversity_not_validated", "source_evaluation_split", "sql_dialect_and_result_equivalence_need_review", "sql_not_executed", "synthetic_reference" ]	false	true	false	task:39fa8f33a9305e900ca13675fb6bac0170cca8782973790a1ceb11768a8eaf9c	1	1	[ "gretel_sql" ]	held_out	curated_candidate	[]	core_candidate	[ "source_sql_complexity:multiple_joins", "sql_structure:selects=2;joins=2;statements=1;advanced=none", "combined_relational_operations_require_calibration" ]	structural_heuristic	Qwen3.6-27B	heuristic_not_model_measured
gretel_sql:ab07b543bfc8fe39af183dfccbfcbbea56c54fdcc2ed059d3fba368b3539092d	task:e01ccf696f0ced08f34001ab5000e2e36722a10436cfc6e17a8287fabcfe8f84	316abd1246b6b05b9070e131b629aa6cfadb0ef348271ebdec2aaae65f94d327	gretel_sql	gretelai/synthetic_text_to_sql	740ab236e64503fba51be1101df7a1be83bf455d	synthetic_text_to_sql_test.snappy.parquet	2368	test	https://huggingface.co/datasets/gretelai/synthetic_text_to_sql/resolve/740ab236e64503fba51be1101df7a1be83bf455d/synthetic_text_to_sql_test.snappy.parquet	apache-2.0	Gretel Synthetic Text-to-SQL dataset creators; source card and author citation preserved in the download manifest.	proposed	Show the total assets under management (AUM) for each risk category.	CREATE TABLE risk_categories (risk_category VARCHAR(20)); INSERT INTO risk_categories (risk_category) VALUES ('Low'), ('Medium'), ('High'); CREATE TABLE client_risk (client_id INT, risk_category VARCHAR(20)); INSERT INTO client_risk (client_id, risk_category) VALUES (1, 'Medium'), (2, 'High'), (3, 'Low');	database_business_operations	database_business_operations.database_query_analysis	heuristic_unreviewed	Mapped from source sql_task_type='analytics and reporting'; no individual SQL execution or semantic review.	en	none	SELECT cr.risk_category, SUM(value) AS total_aum FROM client_risk cr JOIN clients c ON cr.client_id = c.client_id JOIN assets a ON c.client_id = a.client_id GROUP BY cr.risk_category;	{"domain":"financial services","domain_description":"Detailed financial data including investment strategies, risk management, fraud detection, customer analytics, and regulatory compliance.","id":2368,"sql_complexity":"multiple_joins","sql_complexity_description":"two or more joins (specify inner, outer, cross)","sql_...	995ea072daf49f1f0949a26926647727684becf628630bca48de8f7a554ab046	gretel_sql:test:2368	multiple_joins	intermediate	Estimated from source sql_complexity='multiple_joins'; query feature labels are only proxies for difficulty, not measured solver performance. No empirical solver calibration was performed.	heuristic	needs_difficulty_calibration	not_measured	analysis_only_unvalidated	primary	task_annotation	2,367	[ "difficulty_not_calibrated", "not_individually_reviewed", "reference_not_independently_verified", "reference_or_constraint_only_no_rubric", "response_diversity_not_validated", "source_evaluation_split", "sql_dialect_and_result_equivalence_need_review", "sql_not_executed", "synthetic_reference" ]	false	true	false	task:e01ccf696f0ced08f34001ab5000e2e36722a10436cfc6e17a8287fabcfe8f84	1	1	[ "gretel_sql" ]	held_out	curated_candidate	[]	core_candidate	[ "source_sql_complexity:multiple_joins", "sql_structure:selects=1;joins=2;statements=1;advanced=none", "combined_relational_operations_require_calibration" ]	structural_heuristic	Qwen3.6-27B	heuristic_not_model_measured
gretel_sql:cc60b7f88fb56e0122dcfcdb5a09eae47adf79fb115ebf27b4b5c4b739a43073	task:2850724ef74d89b1074046457d0d976e06f5afd1c1cf6c75898959ef931ae3ce	3451c6725ed80d8f22868a4b3d77aa215ec33660bdc22ce05cb774ff5d5023f5	gretel_sql	gretelai/synthetic_text_to_sql	740ab236e64503fba51be1101df7a1be83bf455d	synthetic_text_to_sql_test.snappy.parquet	2403	test	https://huggingface.co/datasets/gretelai/synthetic_text_to_sql/resolve/740ab236e64503fba51be1101df7a1be83bf455d/synthetic_text_to_sql_test.snappy.parquet	apache-2.0	Gretel Synthetic Text-to-SQL dataset creators; source card and author citation preserved in the download manifest.	proposed	Who is the most frequent customer ordering vegetarian items?	CREATE TABLE customers (customer_id INT, customer_name VARCHAR(50)); INSERT INTO customers VALUES (1, 'John Doe'), (2, 'Jane Smith'), (3, 'Alice Johnson'); CREATE TABLE orders (order_id INT, customer_id INT, menu_id INT, order_date DATE); INSERT INTO orders VALUES (1, 1, 1, '2022-01-01'), (2, 2, 3, '2022-01-02'), (3, 3...	database_business_operations	database_business_operations.database_query_analysis	heuristic_unreviewed	Mapped from source sql_task_type='analytics and reporting'; no individual SQL execution or semantic review.	en	none	SELECT customers.customer_name, COUNT(orders.order_id) as order_count FROM customers INNER JOIN orders ON customers.customer_id = orders.customer_id INNER JOIN menu ON orders.menu_id = menu.menu_id WHERE menu.is_vegetarian = true GROUP BY customers.customer_name ORDER BY order_count DESC LIMIT 1;	{"domain":"food services","domain_description":"Menu engineering, customer preferences, inventory management, and sustainability initiatives.","id":2403,"sql_complexity":"multiple_joins","sql_complexity_description":"two or more joins (specify inner, outer, cross)","sql_explanation":"This query retrieves the most frequ...	205be4a4447bec3afb65354d7ae09826797efab257b3ed2c889bb56deb90dcb3	gretel_sql:test:2403	multiple_joins	intermediate	Estimated from source sql_complexity='multiple_joins'; query feature labels are only proxies for difficulty, not measured solver performance. No empirical solver calibration was performed.	heuristic	needs_difficulty_calibration	not_measured	analysis_only_unvalidated	primary	task_annotation	2,402	[ "difficulty_not_calibrated", "not_individually_reviewed", "reference_not_independently_verified", "reference_or_constraint_only_no_rubric", "response_diversity_not_validated", "source_evaluation_split", "sql_dialect_and_result_equivalence_need_review", "sql_not_executed", "synthetic_reference" ]	false	true	false	task:2850724ef74d89b1074046457d0d976e06f5afd1c1cf6c75898959ef931ae3ce	1	1	[ "gretel_sql" ]	held_out	curated_candidate	[]	core_candidate	[ "source_sql_complexity:multiple_joins", "sql_structure:selects=1;joins=2;statements=1;advanced=none", "combined_relational_operations_require_calibration" ]	structural_heuristic	Qwen3.6-27B	heuristic_not_model_measured
gretel_sql:a332634904b0d37f9298ca6c54a94ca29c1b1ab80c8d7d71874a44bb96ec2a7d	task:278722934894ec6169920d8a82b7cf88b0ebf1c7a173c63e1c88e935de4a2eb4	733ded4f64f0576dee5b277c12087d31d9eabfdcddbc351dcc4986a68ad604bc	gretel_sql	gretelai/synthetic_text_to_sql	740ab236e64503fba51be1101df7a1be83bf455d	synthetic_text_to_sql_test.snappy.parquet	2406	test	https://huggingface.co/datasets/gretelai/synthetic_text_to_sql/resolve/740ab236e64503fba51be1101df7a1be83bf455d/synthetic_text_to_sql_test.snappy.parquet	apache-2.0	Gretel Synthetic Text-to-SQL dataset creators; source card and author citation preserved in the download manifest.	proposed	How many days in the last month was the data usage for each customer above the average data usage for that customer?	CREATE TABLE daily_usage (customer_id INT, date DATE, data_usage FLOAT); INSERT INTO daily_usage VALUES (1, '2022-01-01', 5), (1, '2022-01-02', 7);	database_business_operations	database_business_operations.database_query_analysis	heuristic_unreviewed	Mapped from source sql_task_type='analytics and reporting'; no individual SQL execution or semantic review.	en	none	SELECT customer_id, COUNT(*) as days_above_average FROM (SELECT customer_id, date, data_usage, AVG(data_usage) OVER(PARTITION BY customer_id) as avg_usage FROM daily_usage WHERE date >= DATEADD(month, -1, GETDATE())) daily_usage_avg WHERE data_usage > avg_usage GROUP BY customer_id;	{"domain":"telecommunications","domain_description":"Mobile and broadband subscriber data, network infrastructure investments, customer usage patterns, and regulatory compliance information.","id":2406,"sql_complexity":"window functions","sql_complexity_description":"window functions (e.g., ROW_NUMBER, LEAD, LAG, RANk,...	9086022ef6155fc7a76dba8116da00c0bcf249b495784d0f363e6fc84ec54367	gretel_sql:test:2406	window functions	intermediate	Estimated from source sql_complexity='window functions'; query feature labels are only proxies for difficulty, not measured solver performance. No empirical solver calibration was performed.	heuristic	needs_difficulty_calibration	not_measured	analysis_only_unvalidated	primary	task_annotation	2,405	[ "difficulty_not_calibrated", "not_individually_reviewed", "reference_not_independently_verified", "reference_or_constraint_only_no_rubric", "response_diversity_not_validated", "source_evaluation_split", "sql_dialect_and_result_equivalence_need_review", "sql_not_executed", "synthetic_reference" ]	false	true	false	task:278722934894ec6169920d8a82b7cf88b0ebf1c7a173c63e1c88e935de4a2eb4	1	1	[ "gretel_sql" ]	held_out	curated_candidate	[]	core_candidate	[ "source_sql_complexity:window functions", "sql_structure:selects=2;joins=0;statements=1;advanced=OVER", "combined_relational_operations_require_calibration" ]	structural_heuristic	Qwen3.6-27B	heuristic_not_model_measured
gretel_sql:f20f2c48b4e85e44b06760cdd3c072b351c183db07d9caebaa8c4d77b16cbfa7	task:884a5097c84458404db5900b713f5e2ed4e25024badb13815ec91b07992dd08d	ac8f5c7035c4c65a0dabe6474b06c1368394c114e4db075479849e294526d5b7	gretel_sql	gretelai/synthetic_text_to_sql	740ab236e64503fba51be1101df7a1be83bf455d	synthetic_text_to_sql_test.snappy.parquet	2414	test	https://huggingface.co/datasets/gretelai/synthetic_text_to_sql/resolve/740ab236e64503fba51be1101df7a1be83bf455d/synthetic_text_to_sql_test.snappy.parquet	apache-2.0	Gretel Synthetic Text-to-SQL dataset creators; source card and author citation preserved in the download manifest.	proposed	List the policy areas with the lowest citizen satisfaction scores.	CREATE TABLE Policy (Area VARCHAR(20), Score INT); INSERT INTO Policy (Area, Score) VALUES ('Transportation', 70), ('Education', 85), ('Healthcare', 75), ('PublicSafety', 80);	database_business_operations	database_business_operations.database_query_analysis	heuristic_unreviewed	Mapped from source sql_task_type='analytics and reporting'; no individual SQL execution or semantic review.	en	none	SELECT Area, Score FROM (SELECT Area, Score, ROW_NUMBER() OVER (ORDER BY Score) AS RN FROM Policy) X WHERE RN IN (1, 2);	{"domain":"government services","domain_description":"Public service delivery, budget allocation, citizen feedback, and policy impact.","id":2414,"sql_complexity":"window functions","sql_complexity_description":"window functions (e.g., ROW_NUMBER, LEAD, LAG, RANk, NTILE, PERCENT_RANK, etc.) with partitioning and orderi...	12ecf683b9e17c5bb15e342f2461548dfeac01e7fd3e8096b57e9a927016913f	gretel_sql:test:2414	window functions	intermediate	Estimated from source sql_complexity='window functions'; query feature labels are only proxies for difficulty, not measured solver performance. No empirical solver calibration was performed.	heuristic	needs_difficulty_calibration	not_measured	analysis_only_unvalidated	primary	task_annotation	2,413	[ "difficulty_not_calibrated", "not_individually_reviewed", "reference_not_independently_verified", "reference_or_constraint_only_no_rubric", "response_diversity_not_validated", "source_evaluation_split", "sql_dialect_and_result_equivalence_need_review", "sql_not_executed", "synthetic_reference" ]	false	true	false	task:884a5097c84458404db5900b713f5e2ed4e25024badb13815ec91b07992dd08d	1	1	[ "gretel_sql" ]	held_out	curated_candidate	[]	core_candidate	[ "source_sql_complexity:window functions", "sql_structure:selects=2;joins=0;statements=1;advanced=OVER", "combined_relational_operations_require_calibration" ]	structural_heuristic	Qwen3.6-27B	heuristic_not_model_measured
gretel_sql:b5f75a84210e1b0b2bd0670c1bcc7bee161941b8fc60f2f0c567cb4b69ea2880	task:abc96b28f00818e36ff220cff6bbea9941a479c9578f92e2f911973a871cd0da	d2d1e0e8384ab8caa403af0478fc8abbb867207cd6e5a7dce9ba7aac86a0a095	gretel_sql	gretelai/synthetic_text_to_sql	740ab236e64503fba51be1101df7a1be83bf455d	synthetic_text_to_sql_test.snappy.parquet	2552	test	https://huggingface.co/datasets/gretelai/synthetic_text_to_sql/resolve/740ab236e64503fba51be1101df7a1be83bf455d/synthetic_text_to_sql_test.snappy.parquet	apache-2.0	Gretel Synthetic Text-to-SQL dataset creators; source card and author citation preserved in the download manifest.	proposed	How many visitors who identify as 'Male' have spent more than $100 on events in the 'Art' category?	CREATE TABLE Visitors (VisitorID INT, Age INT, Gender VARCHAR(10));CREATE TABLE Events (EventID INT, EventName VARCHAR(20), EventCategory VARCHAR(20));CREATE TABLE VisitorSpending (VisitorID INT, EventID INT, Spending INT);	database_business_operations	database_business_operations.database_query_analysis	heuristic_unreviewed	Mapped from source sql_task_type='analytics and reporting'; no individual SQL execution or semantic review.	en	none	SELECT COUNT(*) AS Num_Visitors FROM Visitors V INNER JOIN VisitorSpending VS ON V.VisitorID = VS.VisitorID INNER JOIN Events E ON VS.EventID = E.EventID WHERE V.Gender = 'Male' AND VS.Spending > 100 AND E.EventCategory = 'Art';	{"domain":"arts and culture","domain_description":"Audience demographics, event attendance, program impact, and funding sources.","id":2552,"sql_complexity":"multiple_joins","sql_complexity_description":"two or more joins (specify inner, outer, cross)","sql_explanation":"The SQL query calculates the number of visitors ...	16098bec0b7cd270c62dfe9df86f17c8c97e47901bd6e463674d87f8ef23cc21	gretel_sql:test:2552	multiple_joins	intermediate	Estimated from source sql_complexity='multiple_joins'; query feature labels are only proxies for difficulty, not measured solver performance. No empirical solver calibration was performed.	heuristic	needs_difficulty_calibration	not_measured	analysis_only_unvalidated	primary	task_annotation	2,551	[ "difficulty_not_calibrated", "not_individually_reviewed", "reference_not_independently_verified", "reference_or_constraint_only_no_rubric", "response_diversity_not_validated", "source_evaluation_split", "sql_dialect_and_result_equivalence_need_review", "sql_not_executed", "synthetic_reference" ]	false	true	false	task:abc96b28f00818e36ff220cff6bbea9941a479c9578f92e2f911973a871cd0da	1	1	[ "gretel_sql" ]	held_out	curated_candidate	[]	core_candidate	[ "source_sql_complexity:multiple_joins", "sql_structure:selects=1;joins=2;statements=1;advanced=none", "combined_relational_operations_require_calibration" ]	structural_heuristic	Qwen3.6-27B	heuristic_not_model_measured
gretel_sql:f447aaf28e68a523761c13757e3ba8852fda2520452925d3ca346463340bb180	task:c0b67bda3f40b05591fa176da5f275d5303107d9b983530792ff4bf1605543d5	fcb79152eef2b1c53a8b9ee174c9f1539cfb6c2303d85e9a44c5f3218bf3496a	gretel_sql	gretelai/synthetic_text_to_sql	740ab236e64503fba51be1101df7a1be83bf455d	synthetic_text_to_sql_test.snappy.parquet	2579	test	https://huggingface.co/datasets/gretelai/synthetic_text_to_sql/resolve/740ab236e64503fba51be1101df7a1be83bf455d/synthetic_text_to_sql_test.snappy.parquet	apache-2.0	Gretel Synthetic Text-to-SQL dataset creators; source card and author citation preserved in the download manifest.	proposed	What is the average number of streams per day for each song by artists from the United States on Apple Music?	CREATE TABLE Artists (ArtistID INT, ArtistName VARCHAR(100), Country VARCHAR(50)); INSERT INTO Artists (ArtistID, ArtistName, Country) VALUES (4, 'Billie Eilish', 'United States'); CREATE TABLE StreamingPlatforms (PlatformID INT, PlatformName VARCHAR(50)); INSERT INTO StreamingPlatforms (PlatformID, PlatformName) VALUE...	database_business_operations	database_business_operations.database_query_analysis	heuristic_unreviewed	Mapped from source sql_task_type='analytics and reporting'; no individual SQL execution or semantic review.	en	none	SELECT ss.SongID, AVG(ss.StreamCount / DATEDIFF('2022-12-31', ss.ReleaseDate)) AS AvgStreamsPerDay FROM SongsStreams ss JOIN Artists a ON ss.ArtistID = a.ArtistID JOIN StreamingPlatforms sp ON ss.PlatformID = sp.PlatformID WHERE a.Country = 'United States' AND sp.PlatformName = 'Apple Music' GROUP BY ss.SongID;	{"domain":"music","domain_description":"Detailed records on music creation, distribution, marketing, and consumption patterns across various genres and platforms.","id":2579,"sql_complexity":"multiple_joins","sql_complexity_description":"two or more joins (specify inner, outer, cross)","sql_explanation":"This query joi...	6606e2008cf362e8ee60b544680b72284d2bbc069d00562dbfe1b2772085530a	gretel_sql:test:2579	multiple_joins	intermediate	Estimated from source sql_complexity='multiple_joins'; query feature labels are only proxies for difficulty, not measured solver performance. No empirical solver calibration was performed.	heuristic	needs_difficulty_calibration	not_measured	analysis_only_unvalidated	primary	task_annotation	2,578	[ "difficulty_not_calibrated", "not_individually_reviewed", "reference_not_independently_verified", "reference_or_constraint_only_no_rubric", "response_diversity_not_validated", "source_evaluation_split", "sql_dialect_and_result_equivalence_need_review", "sql_not_executed", "synthetic_reference" ]	false	true	false	task:c0b67bda3f40b05591fa176da5f275d5303107d9b983530792ff4bf1605543d5	1	1	[ "gretel_sql" ]	held_out	curated_candidate	[]	core_candidate	[ "source_sql_complexity:multiple_joins", "sql_structure:selects=1;joins=2;statements=1;advanced=none", "combined_relational_operations_require_calibration" ]	structural_heuristic	Qwen3.6-27B	heuristic_not_model_measured
gretel_sql:afb6d724f484761d4c88dcf859050dbb00fb68fdc541a2f849f7abfa0b06c1c7	task:344c28431c5da033878145a0f0a90fa8eebefe9c30bbec2c98a030adf043b3b4	590948596d3bfc7c369c040c8ed895083a4b40df4722bb9c57ba7d12cd7649ce	gretel_sql	gretelai/synthetic_text_to_sql	740ab236e64503fba51be1101df7a1be83bf455d	synthetic_text_to_sql_test.snappy.parquet	2592	test	https://huggingface.co/datasets/gretelai/synthetic_text_to_sql/resolve/740ab236e64503fba51be1101df7a1be83bf455d/synthetic_text_to_sql_test.snappy.parquet	apache-2.0	Gretel Synthetic Text-to-SQL dataset creators; source card and author citation preserved in the download manifest.	proposed	What is the third highest account balance for Shariah-compliant finance customers?	CREATE TABLE shariah_compliant_finance(customer_id INT, account_balance DECIMAL(10, 2)); INSERT INTO shariah_compliant_finance VALUES (1, 5000), (2, 7000), (3, 6000), (4, 8000), (5, 9000);	database_business_operations	database_business_operations.database_query_analysis	heuristic_unreviewed	Mapped from source sql_task_type='analytics and reporting'; no individual SQL execution or semantic review.	en	none	SELECT account_balance FROM (SELECT account_balance, RANK() OVER (ORDER BY account_balance DESC) as rn FROM shariah_compliant_finance) t WHERE rn = 3;	{"domain":"finance","domain_description":"Financial data on Shariah-compliant finance, socially responsible lending, financial capability, and financial wellbeing.","id":2592,"sql_complexity":"window functions","sql_complexity_description":"window functions (e.g., ROW_NUMBER, LEAD, LAG, RANk, NTILE, PERCENT_RANK, etc.)...	8a4bb0b8a6b99b04683ce1d5ac43675d6a2f3dde04905b427552be28ed175178	gretel_sql:test:2592	window functions	intermediate	Estimated from source sql_complexity='window functions'; query feature labels are only proxies for difficulty, not measured solver performance. No empirical solver calibration was performed.	heuristic	needs_difficulty_calibration	not_measured	analysis_only_unvalidated	primary	task_annotation	2,591	[ "difficulty_not_calibrated", "not_individually_reviewed", "reference_not_independently_verified", "reference_or_constraint_only_no_rubric", "response_diversity_not_validated", "source_evaluation_split", "sql_dialect_and_result_equivalence_need_review", "sql_not_executed", "synthetic_reference" ]	false	true	false	task:344c28431c5da033878145a0f0a90fa8eebefe9c30bbec2c98a030adf043b3b4	1	1	[ "gretel_sql" ]	held_out	curated_candidate	[]	core_candidate	[ "source_sql_complexity:window functions", "sql_structure:selects=2;joins=0;statements=1;advanced=OVER", "combined_relational_operations_require_calibration" ]	structural_heuristic	Qwen3.6-27B	heuristic_not_model_measured
gretel_sql:f5b826088625277845967b6986835db86dc106452b0c5789fe18a7f4c435f61e	task:4733380b2befc006673446d5c727186547267f3228e1c8965a894a94fc444f94	26258e4b5c0dcf7323a02150ef8c83cce0d4c9e92f0ea3b8017c2aa79715e951	gretel_sql	gretelai/synthetic_text_to_sql	740ab236e64503fba51be1101df7a1be83bf455d	synthetic_text_to_sql_test.snappy.parquet	2615	test	https://huggingface.co/datasets/gretelai/synthetic_text_to_sql/resolve/740ab236e64503fba51be1101df7a1be83bf455d/synthetic_text_to_sql_test.snappy.parquet	apache-2.0	Gretel Synthetic Text-to-SQL dataset creators; source card and author citation preserved in the download manifest.	proposed	What are the total amounts of copper and gold extracted by each company?	CREATE TABLE company (id INT, name VARCHAR(255));CREATE TABLE copper_extraction (company_id INT, amount INT);CREATE TABLE gold_extraction (company_id INT, amount INT);	database_business_operations	database_business_operations.database_query_analysis	heuristic_unreviewed	Mapped from source sql_task_type='analytics and reporting'; no individual SQL execution or semantic review.	en	none	SELECT c.name, SUM(ce.amount) as total_copper, SUM(ge.amount) as total_gold FROM company c LEFT JOIN copper_extraction ce ON c.id = ce.company_id LEFT JOIN gold_extraction ge ON c.id = ge.company_id GROUP BY c.name;	{"domain":"mining industry","domain_description":"Mineral extraction statistics, environmental impact assessments, labor productivity metrics, and geological survey information.","id":2615,"sql_complexity":"multiple_joins","sql_complexity_description":"two or more joins (specify inner, outer, cross)","sql_explanation":...	135bb1468281c3f7cfee4c6e039c37918c29ee576c1f322096bf58e603081e37	gretel_sql:test:2615	multiple_joins	intermediate	Estimated from source sql_complexity='multiple_joins'; query feature labels are only proxies for difficulty, not measured solver performance. No empirical solver calibration was performed.	heuristic	needs_difficulty_calibration	not_measured	analysis_only_unvalidated	primary	task_annotation	2,614	[ "difficulty_not_calibrated", "not_individually_reviewed", "reference_not_independently_verified", "reference_or_constraint_only_no_rubric", "response_diversity_not_validated", "source_evaluation_split", "sql_dialect_and_result_equivalence_need_review", "sql_not_executed", "synthetic_reference" ]	false	true	false	task:4733380b2befc006673446d5c727186547267f3228e1c8965a894a94fc444f94	1	1	[ "gretel_sql" ]	held_out	curated_candidate	[]	core_candidate	[ "source_sql_complexity:multiple_joins", "sql_structure:selects=1;joins=2;statements=1;advanced=none", "combined_relational_operations_require_calibration" ]	structural_heuristic	Qwen3.6-27B	heuristic_not_model_measured
gretel_sql:b85e090738ee436463fd45c56a7810ff2eec7a3790ef54ee4268b2d7e0b22fb1	task:8d2a067fb29216c559bdd8864e70191bee912532808b0db168dc4fc317b770e3	9476e8d6f8dbe7573318bbd9d235b4491720d585a41c5bf87fb3eb310d5dc5b1	gretel_sql	gretelai/synthetic_text_to_sql	740ab236e64503fba51be1101df7a1be83bf455d	synthetic_text_to_sql_test.snappy.parquet	2656	test	https://huggingface.co/datasets/gretelai/synthetic_text_to_sql/resolve/740ab236e64503fba51be1101df7a1be83bf455d/synthetic_text_to_sql_test.snappy.parquet	apache-2.0	Gretel Synthetic Text-to-SQL dataset creators; source card and author citation preserved in the download manifest.	proposed	Find the drugs and their respective total sales for rare diseases indication with sales greater than the average sales for infectious diseases.	CREATE TABLE sales (id INT, drug_id INT, quarter INT, year INT, revenue FLOAT); INSERT INTO sales (id, drug_id, quarter, year, revenue) VALUES (1, 1, 1, 2022, 1500000); CREATE TABLE drugs (id INT, name VARCHAR(50), company VARCHAR(50), indication VARCHAR(50)); INSERT INTO drugs (id, name, company, indication) VALUES (1...	database_business_operations	database_business_operations.database_query_analysis	heuristic_unreviewed	Mapped from source sql_task_type='analytics and reporting'; no individual SQL execution or semantic review.	en	none	SELECT s.drug_id, d.name, SUM(s.revenue) as total_sales FROM sales s JOIN drugs d ON s.drug_id = d.id WHERE d.indication = 'Rare_Diseases' GROUP BY s.drug_id HAVING total_sales > (SELECT AVG(s2.revenue) FROM sales s2 JOIN drugs d2 ON s2.drug_id = d2.id WHERE d2.indication = 'Infectious_Diseases')	{"domain":"pharmaceuticals","domain_description":"Clinical trial outcomes, drug approval data, sales figures, R&D expenditures, and market access strategies.","id":2656,"sql_complexity":"multiple_joins","sql_complexity_description":"two or more joins (specify inner, outer, cross)","sql_explanation":"This query calculat...	e549e71ecb3c8392dd4e94877df52a9821848100c687669e5d74dcefc46482da	gretel_sql:test:2656	multiple_joins	intermediate	Estimated from source sql_complexity='multiple_joins'; query feature labels are only proxies for difficulty, not measured solver performance. No empirical solver calibration was performed.	heuristic	needs_difficulty_calibration	not_measured	analysis_only_unvalidated	primary	task_annotation	2,655	[ "difficulty_not_calibrated", "not_individually_reviewed", "reference_not_independently_verified", "reference_or_constraint_only_no_rubric", "response_diversity_not_validated", "source_evaluation_split", "sql_dialect_and_result_equivalence_need_review", "sql_not_executed", "synthetic_reference" ]	false	true	false	task:8d2a067fb29216c559bdd8864e70191bee912532808b0db168dc4fc317b770e3	1	1	[ "gretel_sql" ]	held_out	curated_candidate	[]	core_candidate	[ "source_sql_complexity:multiple_joins", "sql_structure:selects=2;joins=2;statements=1;advanced=none", "combined_relational_operations_require_calibration" ]	structural_heuristic	Qwen3.6-27B	heuristic_not_model_measured
gretel_sql:ac9d5f32866a2cc57aea2d8a2910e96ed94d965d39087354e0e7a4b74bb49e8d	task:a5dffb355a3627f86f4c1e6f23251cbc0c61fed24f93fb6e3d72bbc0f98ee6d1	2f170ea2dae6a762e028fd6b98b5b645a72f2b5580e98f2fb50076b032ad0e45	gretel_sql	gretelai/synthetic_text_to_sql	740ab236e64503fba51be1101df7a1be83bf455d	synthetic_text_to_sql_test.snappy.parquet	2748	test	https://huggingface.co/datasets/gretelai/synthetic_text_to_sql/resolve/740ab236e64503fba51be1101df7a1be83bf455d/synthetic_text_to_sql_test.snappy.parquet	apache-2.0	Gretel Synthetic Text-to-SQL dataset creators; source card and author citation preserved in the download manifest.	proposed	Which regions had the highest number of first-time visitors to the theatre in 2022?	CREATE TABLE TheatreVisits (ID INT, VisitDate DATE, VisitorID INT, Region VARCHAR(50)); CREATE TABLE FirstTimeVisitors (ID INT, VisitorID INT, FirstVisit DATE);	database_business_operations	database_business_operations.database_query_analysis	heuristic_unreviewed	Mapped from source sql_task_type='analytics and reporting'; no individual SQL execution or semantic review.	en	none	SELECT r.Region, COUNT(DISTINCT v.VisitorID) as FirstTimeVisitorsCount FROM TheatreVisits v JOIN FirstTimeVisitors f ON v.VisitorID = f.VisitorID JOIN (SELECT DISTINCT Region FROM TheatreVisits WHERE VisitDate >= '2022-01-01' AND VisitDate < '2023-01-01') r ON v.Region = r.Region WHERE f.FirstVisit >= '2022-01-01' AND ...	{"domain":"arts and culture","domain_description":"Audience demographics, event attendance, program impact, and funding sources.","id":2748,"sql_complexity":"multiple_joins","sql_complexity_description":"two or more joins (specify inner, outer, cross)","sql_explanation":"This SQL query identifies the regions with the h...	81abab23f8d15281e5225f602f0568c795aa98d3b5770fa81ce71dfd9f45a616	gretel_sql:test:2748	multiple_joins	intermediate	Estimated from source sql_complexity='multiple_joins'; query feature labels are only proxies for difficulty, not measured solver performance. No empirical solver calibration was performed.	heuristic	needs_difficulty_calibration	not_measured	analysis_only_unvalidated	primary	task_annotation	2,747	[ "difficulty_not_calibrated", "not_individually_reviewed", "reference_not_independently_verified", "reference_or_constraint_only_no_rubric", "response_diversity_not_validated", "source_evaluation_split", "sql_dialect_and_result_equivalence_need_review", "sql_not_executed", "synthetic_reference" ]	false	true	false	task:a5dffb355a3627f86f4c1e6f23251cbc0c61fed24f93fb6e3d72bbc0f98ee6d1	1	1	[ "gretel_sql" ]	held_out	curated_candidate	[]	core_candidate	[ "source_sql_complexity:multiple_joins", "sql_structure:selects=2;joins=2;statements=1;advanced=none", "combined_relational_operations_require_calibration" ]	structural_heuristic	Qwen3.6-27B	heuristic_not_model_measured
gretel_sql:f7afe22c98f23ff48611757404f8cf5ce4cffbfd9897eb62c9042e2f7207c4c0	task:65b3bdfae15277b84e9df16f25e2b49151f67a2ba57c5ce4204e559bb901f229	cb9917c0f38830ea8f9ed1ace13845384380fe117e8f4ddfc83975479c3f4ef2	gretel_sql	gretelai/synthetic_text_to_sql	740ab236e64503fba51be1101df7a1be83bf455d	synthetic_text_to_sql_test.snappy.parquet	2812	test	https://huggingface.co/datasets/gretelai/synthetic_text_to_sql/resolve/740ab236e64503fba51be1101df7a1be83bf455d/synthetic_text_to_sql_test.snappy.parquet	apache-2.0	Gretel Synthetic Text-to-SQL dataset creators; source card and author citation preserved in the download manifest.	proposed	What is the total number of volunteers and total volunteer hours for each region, excluding the top 3 regions with the highest total volunteer hours?	CREATE TABLE Volunteers (VolunteerID INT, VolunteerName TEXT, Region TEXT, VolunteerHours INT, EventDate DATE); INSERT INTO Volunteers VALUES (1, 'Ahmed Al-Hassan', 'Middle East and North Africa', 20, '2022-01-01'), (2, 'Fatima Al-Farsi', 'Europe', 15, '2022-02-01');	database_business_operations	database_business_operations.database_query_analysis	heuristic_unreviewed	Mapped from source sql_task_type='analytics and reporting'; no individual SQL execution or semantic review.	en	none	SELECT Region, COUNT(VolunteerID) as TotalVolunteers, SUM(VolunteerHours) as TotalHours FROM Volunteers v JOIN (SELECT Region, ROW_NUMBER() OVER (ORDER BY SUM(VolunteerHours) DESC) as rn FROM Volunteers GROUP BY Region) tmp ON v.Region = tmp.Region WHERE rn > 3 GROUP BY Region;	{"domain":"nonprofit","domain_description":"Nonprofit data on charitable giving trends, social impact assessment, capacity building, and volunteer management.","id":2812,"sql_complexity":"window functions","sql_complexity_description":"window functions (e.g., ROW_NUMBER, LEAD, LAG, RANk, NTILE, PERCENT_RANK, etc.) with...	f47c8fdbb70476ee7f6e735b33a0de19e8e252620ff432823e01af8f49dd932c	gretel_sql:test:2812	window functions	intermediate	Estimated from source sql_complexity='window functions'; query feature labels are only proxies for difficulty, not measured solver performance. No empirical solver calibration was performed.	heuristic	needs_difficulty_calibration	not_measured	analysis_only_unvalidated	primary	task_annotation	2,811	[ "difficulty_not_calibrated", "not_individually_reviewed", "reference_not_independently_verified", "reference_or_constraint_only_no_rubric", "response_diversity_not_validated", "source_evaluation_split", "sql_dialect_and_result_equivalence_need_review", "sql_not_executed", "synthetic_reference" ]	false	true	false	task:65b3bdfae15277b84e9df16f25e2b49151f67a2ba57c5ce4204e559bb901f229	1	1	[ "gretel_sql" ]	held_out	curated_candidate	[]	core_candidate	[ "source_sql_complexity:window functions", "sql_structure:selects=2;joins=1;statements=1;advanced=OVER", "combined_relational_operations_require_calibration" ]	structural_heuristic	Qwen3.6-27B	heuristic_not_model_measured
gretel_sql:e4d918095ddfa6fc725c7955940802c92f22387cbc94b332bcdd8e394ec7cdab	task:420c9f08d33854580cde8f67df674be84d84f6632ac55692896afa7d2170b631	17c05378d8f8cf40a3432d1110d2ff3603d0b877a232038462fce5ac4e3836e7	gretel_sql	gretelai/synthetic_text_to_sql	740ab236e64503fba51be1101df7a1be83bf455d	synthetic_text_to_sql_test.snappy.parquet	2897	test	https://huggingface.co/datasets/gretelai/synthetic_text_to_sql/resolve/740ab236e64503fba51be1101df7a1be83bf455d/synthetic_text_to_sql_test.snappy.parquet	apache-2.0	Gretel Synthetic Text-to-SQL dataset creators; source card and author citation preserved in the download manifest.	proposed	Calculate the percentage of mobile and broadband subscribers in each region.	CREATE TABLE mobile_subscribers (subscriber_id INT, region_id INT); INSERT INTO mobile_subscribers (subscriber_id, region_id) VALUES (1, 1), (2, 2), (3, 3), (4, 4), (5, 1), (6, 2), (7, 3), (8, 4); CREATE TABLE broadband_subscribers (subscriber_id INT, region_id INT); INSERT INTO broadband_subscribers (subscriber_id, re...	database_business_operations	database_business_operations.database_query_analysis	heuristic_unreviewed	Mapped from source sql_task_type='analytics and reporting'; no individual SQL execution or semantic review.	en	none	SELECT r.region_name, (COUNT(m.subscriber_id) * 100.0 / (COUNT(m.subscriber_id) + COUNT(b.subscriber_id))) AS mobile_percentage, (COUNT(b.subscriber_id) * 100.0 / (COUNT(m.subscriber_id) + COUNT(b.subscriber_id))) AS broadband_percentage FROM regions r LEFT JOIN mobile_subscribers m ON r.region_id = m.region_id LEFT JO...	{"domain":"telecommunications","domain_description":"Mobile and broadband subscriber data, network infrastructure investments, customer usage patterns, and regulatory compliance information.","id":2897,"sql_complexity":"multiple_joins","sql_complexity_description":"two or more joins (specify inner, outer, cross)","sql_...	1d4a470a3ec111ed13317b4315a2030c890675627ab5f3a38ca0c1f7a4d534a8	gretel_sql:test:2897	multiple_joins	intermediate	Estimated from source sql_complexity='multiple_joins'; query feature labels are only proxies for difficulty, not measured solver performance. No empirical solver calibration was performed.	heuristic	needs_difficulty_calibration	not_measured	analysis_only_unvalidated	primary	task_annotation	2,896	[ "difficulty_not_calibrated", "not_individually_reviewed", "reference_not_independently_verified", "reference_or_constraint_only_no_rubric", "response_diversity_not_validated", "source_evaluation_split", "sql_dialect_and_result_equivalence_need_review", "sql_not_executed", "synthetic_reference" ]	false	true	false	task:420c9f08d33854580cde8f67df674be84d84f6632ac55692896afa7d2170b631	1	1	[ "gretel_sql" ]	held_out	curated_candidate	[]	core_candidate	[ "source_sql_complexity:multiple_joins", "sql_structure:selects=1;joins=2;statements=1;advanced=none", "combined_relational_operations_require_calibration" ]	structural_heuristic	Qwen3.6-27B	heuristic_not_model_measured
gretel_sql:2874010772bb6c4e99b9b82ef8be374fc95d4e2118f55e1f57de888eb01de3d6	task:3318ed6f9f831d341e534823135017c335aa3a686e9500a89ae4adb7ca65283e	fec8f6634ada1021f86b1286127345c681daf99b0b5a311818527485999fe463	gretel_sql	gretelai/synthetic_text_to_sql	740ab236e64503fba51be1101df7a1be83bf455d	synthetic_text_to_sql_test.snappy.parquet	2914	test	https://huggingface.co/datasets/gretelai/synthetic_text_to_sql/resolve/740ab236e64503fba51be1101df7a1be83bf455d/synthetic_text_to_sql_test.snappy.parquet	apache-2.0	Gretel Synthetic Text-to-SQL dataset creators; source card and author citation preserved in the download manifest.	proposed	Find the top 5 most profitable garments based on their sales in Q2 of 2022, and display the garment_name and total revenue.	CREATE TABLE Garments (garment_id INT, garment_name VARCHAR(50), category VARCHAR(50)); INSERT INTO Garments (garment_id, garment_name, category) VALUES (1, 'Cotton T-Shirt', 'Tops'), (2, 'Jeans', 'Bottoms'), (3, 'Silk Blouse', 'Tops'), (4, 'Wool Coat', 'Outerwear'); CREATE TABLE Sales_Details (sale_id INT, garment_id ...	database_business_operations	database_business_operations.database_query_analysis	heuristic_unreviewed	Mapped from source sql_task_type='analytics and reporting'; no individual SQL execution or semantic review.	en	none	SELECT g.garment_name, SUM(sd.sale_quantity * sd.sale_price) AS total_revenue FROM Sales s JOIN Sales_Details sd ON s.sale_id = sd.sale_id JOIN Garments g ON sd.garment_id = g.garment_id WHERE s.sale_date BETWEEN '2022-04-01' AND '2022-06-30' GROUP BY g.garment_name ORDER BY total_revenue DESC LIMIT 5;	{"domain":"fashion","domain_description":"Trend forecasting data, garment manufacturing information, retail sales statistics, and sustainability metrics in the industry.","id":2914,"sql_complexity":"multiple_joins","sql_complexity_description":"two or more joins (specify inner, outer, cross)","sql_explanation":"The SQL...	8e2b72b187f97bad98c3bfd04a7f5a9bc2270410a2636b3f670f97ff8206e77a	gretel_sql:test:2914	multiple_joins	intermediate	Estimated from source sql_complexity='multiple_joins'; query feature labels are only proxies for difficulty, not measured solver performance. No empirical solver calibration was performed.	heuristic	needs_difficulty_calibration	not_measured	analysis_only_unvalidated	primary	task_annotation	2,913	[ "difficulty_not_calibrated", "not_individually_reviewed", "reference_not_independently_verified", "reference_or_constraint_only_no_rubric", "response_diversity_not_validated", "source_evaluation_split", "sql_dialect_and_result_equivalence_need_review", "sql_not_executed", "synthetic_reference" ]	false	true	false	task:3318ed6f9f831d341e534823135017c335aa3a686e9500a89ae4adb7ca65283e	1	1	[ "gretel_sql" ]	held_out	curated_candidate	[]	core_candidate	[ "source_sql_complexity:multiple_joins", "sql_structure:selects=1;joins=2;statements=1;advanced=none", "combined_relational_operations_require_calibration" ]	structural_heuristic	Qwen3.6-27B	heuristic_not_model_measured
gretel_sql:1ceadb816d4df8b574d97033bb7705776acc5a3126123cdb95e7280917e968ed	task:7e621f3f72142f2cf657492293b53f62eab3cd0d8889e33ea3bc83b0644529ad	dc3b60d2d60dcc2939da8133faff00e09a9a67f4f25d22b620b21f0f1d9c5173	gretel_sql	gretelai/synthetic_text_to_sql	740ab236e64503fba51be1101df7a1be83bf455d	synthetic_text_to_sql_test.snappy.parquet	2986	test	https://huggingface.co/datasets/gretelai/synthetic_text_to_sql/resolve/740ab236e64503fba51be1101df7a1be83bf455d/synthetic_text_to_sql_test.snappy.parquet	apache-2.0	Gretel Synthetic Text-to-SQL dataset creators; source card and author citation preserved in the download manifest.	proposed	What is the percentage of workers who identify as female or male in each department, including the total number of workers in each department?	CREATE TABLE department (dept_id INT, dept_name VARCHAR(50), worker_id INT); INSERT INTO department (dept_id, dept_name, worker_id) VALUES (1, 'Mining', 1), (1, 'Mining', 5), (2, 'Engineering', 2); CREATE TABLE worker_demographics (worker_id INT, worker_gender VARCHAR(10)); INSERT INTO worker_demographics (worker_id, w...	database_business_operations	database_business_operations.database_query_analysis	heuristic_unreviewed	Mapped from source sql_task_type='analytics and reporting'; no individual SQL execution or semantic review.	en	none	SELECT dept_name, worker_gender, COUNT() as count, COUNT() * 100.0 / (SELECT COUNT() FROM department d JOIN worker_demographics w ON d.worker_id = w.worker_id) as percentage, (SELECT COUNT() FROM department d JOIN worker_demographics w ON d.worker_id = w.worker_id) as total FROM department d JOIN worker_demographic...	{"domain":"mining operations","domain_description":"Mining operation data, environmental impact stats, workforce diversity, and resource depletion metrics.","id":2986,"sql_complexity":"multiple_joins","sql_complexity_description":"two or more joins (specify inner, outer, cross)","sql_explanation":"The SQL query calcula...	5cf29cc22a18fb7e00c2158beea55387bddf121ad7f6ce995107cbab02bb6784	gretel_sql:test:2986	multiple_joins	intermediate	Estimated from source sql_complexity='multiple_joins'; query feature labels are only proxies for difficulty, not measured solver performance. No empirical solver calibration was performed.	heuristic	needs_difficulty_calibration	not_measured	analysis_only_unvalidated	primary	task_annotation	2,985	[ "difficulty_not_calibrated", "not_individually_reviewed", "reference_not_independently_verified", "reference_or_constraint_only_no_rubric", "response_diversity_not_validated", "source_evaluation_split", "sql_dialect_and_result_equivalence_need_review", "sql_not_executed", "synthetic_reference" ]	false	true	false	task:7e621f3f72142f2cf657492293b53f62eab3cd0d8889e33ea3bc83b0644529ad	1	1	[ "gretel_sql" ]	held_out	curated_candidate	[]	core_candidate	[ "source_sql_complexity:multiple_joins", "sql_structure:selects=3;joins=3;statements=1;advanced=none", "combined_relational_operations_require_calibration" ]	structural_heuristic	Qwen3.6-27B	heuristic_not_model_measured
gretel_sql:1e0fc403ba78ac831e1b18fcab279833be173f3d11b5e6015d71ff20abc2a5f2	task:9c5e50a8080a8f6466398ea68ebc0491b21d09fec029cb7e4583eaba32ce7637	a97c7808766c8ad2f4a6b628c106a16bbf3fc20a201c39018aba76ee343cb486	gretel_sql	gretelai/synthetic_text_to_sql	740ab236e64503fba51be1101df7a1be83bf455d	synthetic_text_to_sql_test.snappy.parquet	2990	test	https://huggingface.co/datasets/gretelai/synthetic_text_to_sql/resolve/740ab236e64503fba51be1101df7a1be83bf455d/synthetic_text_to_sql_test.snappy.parquet	apache-2.0	Gretel Synthetic Text-to-SQL dataset creators; source card and author citation preserved in the download manifest.	proposed	What is the sum of donations and the number of volunteers for organizations that have more than 2 volunteers?	CREATE TABLE Organization (OrgID INT, Name VARCHAR(255)); CREATE TABLE Volunteer (VolID INT, OrgID INT, Name VARCHAR(255)); CREATE TABLE Donation (DonID INT, OrgID INT, Amount DECIMAL(10, 2)); INSERT INTO Organization VALUES (1, 'Greenpeace'), (2, 'Red Cross'), (3, 'Wildlife Fund'); INSERT INTO Volunteer VALUES (1, 1, ...	database_business_operations	database_business_operations.database_query_analysis	heuristic_unreviewed	Mapped from source sql_task_type='analytics and reporting'; no individual SQL execution or semantic review.	en	none	SELECT o.OrgID, SUM(d.Amount) as TotalDonations, COUNT(v.VolID) as TotalVolunteers FROM Organization o LEFT JOIN Volunteer v ON o.OrgID = v.OrgID LEFT JOIN Donation d ON o.OrgID = d.OrgID WHERE (SELECT COUNT(*) FROM Volunteer WHERE Volunteer.OrgID = o.OrgID) > 2 GROUP BY o.OrgID;	{"domain":"nonprofit","domain_description":"Nonprofit data on charitable giving trends, social impact assessment, capacity building, and volunteer management.","id":2990,"sql_complexity":"multiple_joins","sql_complexity_description":"two or more joins (specify inner, outer, cross)","sql_explanation":"This SQL query use...	5baa9fbaa4e1c6b4c2ddcf096ea1e82b3bd9fc4a99f99c80ea90a5e63d09bc2b	gretel_sql:test:2990	multiple_joins	intermediate	Estimated from source sql_complexity='multiple_joins'; query feature labels are only proxies for difficulty, not measured solver performance. No empirical solver calibration was performed.	heuristic	needs_difficulty_calibration	not_measured	analysis_only_unvalidated	primary	task_annotation	2,989	[ "difficulty_not_calibrated", "not_individually_reviewed", "reference_not_independently_verified", "reference_or_constraint_only_no_rubric", "response_diversity_not_validated", "source_evaluation_split", "sql_dialect_and_result_equivalence_need_review", "sql_not_executed", "synthetic_reference" ]	false	true	false	task:9c5e50a8080a8f6466398ea68ebc0491b21d09fec029cb7e4583eaba32ce7637	1	1	[ "gretel_sql" ]	held_out	curated_candidate	[]	core_candidate	[ "source_sql_complexity:multiple_joins", "sql_structure:selects=2;joins=2;statements=1;advanced=none", "combined_relational_operations_require_calibration" ]	structural_heuristic	Qwen3.6-27B	heuristic_not_model_measured
gretel_sql:6f5e1f1be0cc8ebf4e96f142183c0f4e728a4e07863e06efa3978c0ec0937959	task:a1c3110c35545b3ab287245d48d8d2c8275501221e9d52e76940d1f6262d80d7	eb74d1215e3d2271344366d8f911d4363b3f646fdbc7d2b87d8c81784e0c4724	gretel_sql	gretelai/synthetic_text_to_sql	740ab236e64503fba51be1101df7a1be83bf455d	synthetic_text_to_sql_test.snappy.parquet	3002	test	https://huggingface.co/datasets/gretelai/synthetic_text_to_sql/resolve/740ab236e64503fba51be1101df7a1be83bf455d/synthetic_text_to_sql_test.snappy.parquet	apache-2.0	Gretel Synthetic Text-to-SQL dataset creators; source card and author citation preserved in the download manifest.	proposed	Get the product with the highest price from each supplier.	CREATE TABLE product (product_id INT, name VARCHAR(255), price DECIMAL(5,2), supplier_id INT); INSERT INTO product (product_id, name, price, supplier_id) VALUES (1, 'Organic Cotton T-Shirt', 20.99, 1), (2, 'Polyester Hoodie', 35.99, 2), (3, 'Bamboo Socks', 9.99, 1);	database_business_operations	database_business_operations.database_query_analysis	heuristic_unreviewed	Mapped from source sql_task_type='analytics and reporting'; no individual SQL execution or semantic review.	en	none	SELECT product_id, name, price, supplier_id FROM (SELECT product_id, name, price, supplier_id, RANK() OVER (PARTITION BY supplier_id ORDER BY price DESC) AS tier FROM product) AS tiered_products WHERE tier = 1;	{"domain":"retail","domain_description":"Retail data on circular supply chains, ethical labor practices, product transparency, and consumer education.","id":3002,"sql_complexity":"window functions","sql_complexity_description":"window functions (e.g., ROW_NUMBER, LEAD, LAG, RANk, NTILE, PERCENT_RANK, etc.) with partiti...	ed6383d8dfce993056019304cb98dbc2acfc992b6ce6435237659b0cd92e58b1	gretel_sql:test:3002	window functions	intermediate	Estimated from source sql_complexity='window functions'; query feature labels are only proxies for difficulty, not measured solver performance. No empirical solver calibration was performed.	heuristic	needs_difficulty_calibration	not_measured	analysis_only_unvalidated	primary	task_annotation	3,001	[ "difficulty_not_calibrated", "not_individually_reviewed", "reference_not_independently_verified", "reference_or_constraint_only_no_rubric", "response_diversity_not_validated", "source_evaluation_split", "sql_dialect_and_result_equivalence_need_review", "sql_not_executed", "synthetic_reference" ]	false	true	false	task:a1c3110c35545b3ab287245d48d8d2c8275501221e9d52e76940d1f6262d80d7	1	1	[ "gretel_sql" ]	held_out	curated_candidate	[]	core_candidate	[ "source_sql_complexity:window functions", "sql_structure:selects=2;joins=0;statements=1;advanced=OVER", "combined_relational_operations_require_calibration" ]	structural_heuristic	Qwen3.6-27B	heuristic_not_model_measured
gretel_sql:076e966c6bf65c7d758ebcda1b05260275263b73b0a5e2e66a06acb673278602	task:ba5e3fce5e558efc7dd2e06870ca5d5afd85a22601e86082ed697245dc58a328	8cf9ee7742fdd5cde01fc361807693b93a375e4fef8918c5a3c487fd4f2afe92	gretel_sql	gretelai/synthetic_text_to_sql	740ab236e64503fba51be1101df7a1be83bf455d	synthetic_text_to_sql_test.snappy.parquet	3012	test	https://huggingface.co/datasets/gretelai/synthetic_text_to_sql/resolve/740ab236e64503fba51be1101df7a1be83bf455d/synthetic_text_to_sql_test.snappy.parquet	apache-2.0	Gretel Synthetic Text-to-SQL dataset creators; source card and author citation preserved in the download manifest.	proposed	List the top 2 states with the lowest wastewater treatment plant efficiency in ascending order.	CREATE TABLE WastewaterTreatmentPlants (ID INT, State VARCHAR(20), Efficiency FLOAT); INSERT INTO WastewaterTreatmentPlants (ID, State, Efficiency) VALUES (7, 'New York', 0.85), (8, 'New York', 0.87), (9, 'Pennsylvania', 0.83), (10, 'Pennsylvania', 0.84);	database_business_operations	database_business_operations.database_query_analysis	heuristic_unreviewed	Mapped from source sql_task_type='analytics and reporting'; no individual SQL execution or semantic review.	en	none	SELECT State, Efficiency FROM (SELECT State, Efficiency, ROW_NUMBER() OVER (ORDER BY Efficiency ASC) as rn FROM WastewaterTreatmentPlants) tmp WHERE rn <= 2	{"domain":"water resources","domain_description":"Water usage metrics, drought impact assessments, wastewater treatment data, and water conservation initiatives.","id":3012,"sql_complexity":"window functions","sql_complexity_description":"window functions (e.g., ROW_NUMBER, LEAD, LAG, RANk, NTILE, PERCENT_RANK, etc.) w...	0d84d914111aabeff77ace870c741331565ad74e608d34af05a819a4cb0be698	gretel_sql:test:3012	window functions	intermediate	Estimated from source sql_complexity='window functions'; query feature labels are only proxies for difficulty, not measured solver performance. No empirical solver calibration was performed.	heuristic	needs_difficulty_calibration	not_measured	analysis_only_unvalidated	primary	task_annotation	3,011	[ "difficulty_not_calibrated", "not_individually_reviewed", "reference_not_independently_verified", "reference_or_constraint_only_no_rubric", "response_diversity_not_validated", "source_evaluation_split", "sql_dialect_and_result_equivalence_need_review", "sql_not_executed", "synthetic_reference" ]	false	true	false	task:ba5e3fce5e558efc7dd2e06870ca5d5afd85a22601e86082ed697245dc58a328	1	1	[ "gretel_sql" ]	held_out	curated_candidate	[]	core_candidate	[ "source_sql_complexity:window functions", "sql_structure:selects=2;joins=0;statements=1;advanced=OVER", "combined_relational_operations_require_calibration" ]	structural_heuristic	Qwen3.6-27B	heuristic_not_model_measured
gretel_sql:6953382541bca296a015695204a92c56cb76748d8bb12d04c5f06fe96dc96b16	task:d08abe74b58afa184f438f5ce47ab4022d7730a773944373c419ac0927ade85d	30703406ac3353a9211afa741c795d9e914958aac0f2691090afde8477d8a4bf	gretel_sql	gretelai/synthetic_text_to_sql	740ab236e64503fba51be1101df7a1be83bf455d	synthetic_text_to_sql_test.snappy.parquet	3052	test	https://huggingface.co/datasets/gretelai/synthetic_text_to_sql/resolve/740ab236e64503fba51be1101df7a1be83bf455d/synthetic_text_to_sql_test.snappy.parquet	apache-2.0	Gretel Synthetic Text-to-SQL dataset creators; source card and author citation preserved in the download manifest.	proposed	What is the total revenue for each route type, by day?	CREATE TABLE Routes (RouteID int, RouteType varchar(10), StartingLocation varchar(20)); CREATE TABLE Fares (RouteID int, Fare float); CREATE TABLE PassengerCounts (RouteID int, Date date, PassengerCount int); INSERT INTO Routes VALUES (1, 'Bus', 'City Center'), (2, 'Tram', 'City Center'); INSERT INTO Fares VALUES (1, 2...	database_business_operations	database_business_operations.database_query_analysis	heuristic_unreviewed	Mapped from source sql_task_type='analytics and reporting'; no individual SQL execution or semantic review.	en	none	SELECT Routes.RouteType, PassengerCounts.Date, SUM(Fares.Fare * PassengerCounts.PassengerCount) as daily_revenue FROM Routes INNER JOIN Fares ON Routes.RouteID = Fares.RouteID INNER JOIN PassengerCounts ON Routes.RouteID = PassengerCounts.RouteID GROUP BY Routes.RouteType, PassengerCounts.Date;	{"domain":"public transportation","domain_description":"Extensive data on route planning, fare collection, vehicle maintenance, and accessibility in public transportation.","id":3052,"sql_complexity":"multiple_joins","sql_complexity_description":"two or more joins (specify inner, outer, cross)","sql_explanation":"This ...	aa97728903a8794b251640758c990fb3ab859ba58155ab4f095e6756fde06c40	gretel_sql:test:3052	multiple_joins	intermediate	Estimated from source sql_complexity='multiple_joins'; query feature labels are only proxies for difficulty, not measured solver performance. No empirical solver calibration was performed.	heuristic	needs_difficulty_calibration	not_measured	analysis_only_unvalidated	primary	task_annotation	3,051	[ "difficulty_not_calibrated", "not_individually_reviewed", "reference_not_independently_verified", "reference_or_constraint_only_no_rubric", "response_diversity_not_validated", "source_evaluation_split", "sql_dialect_and_result_equivalence_need_review", "sql_not_executed", "synthetic_reference" ]	false	true	false	task:d08abe74b58afa184f438f5ce47ab4022d7730a773944373c419ac0927ade85d	1	1	[ "gretel_sql" ]	held_out	curated_candidate	[]	core_candidate	[ "source_sql_complexity:multiple_joins", "sql_structure:selects=1;joins=2;statements=1;advanced=none", "combined_relational_operations_require_calibration" ]	structural_heuristic	Qwen3.6-27B	heuristic_not_model_measured
gretel_sql:17075010b3cadb8cdec64ff60acb6e420a52bfb5535830b2996777b10880f3c6	task:b0f0e994af3b7aa8346219cc6b4745e38c69c75b1b3a845b9c7bac724ef85718	0a2bcc6c594041f018570c74c293bde47d7cfafeaa8680ae844e985b1cea34f4	gretel_sql	gretelai/synthetic_text_to_sql	740ab236e64503fba51be1101df7a1be83bf455d	synthetic_text_to_sql_test.snappy.parquet	3074	test	https://huggingface.co/datasets/gretelai/synthetic_text_to_sql/resolve/740ab236e64503fba51be1101df7a1be83bf455d/synthetic_text_to_sql_test.snappy.parquet	apache-2.0	Gretel Synthetic Text-to-SQL dataset creators; source card and author citation preserved in the download manifest.	proposed	Find the top 5 players with the highest scores in the 'historical_tournaments' view, including their scores and the names of the tournaments they participated in.	CREATE VIEW historical_tournaments AS SELECT tournaments.tournament_name, players.player_name, players.score FROM tournaments JOIN players_scores ON tournaments.tournament_id = players_scores.tournament_id JOIN players ON players_scores.player_id = players.player_id; CREATE TABLE tournaments (tournament_id INT, tourna...	database_business_operations	database_business_operations.database_query_analysis	heuristic_unreviewed	Mapped from source sql_task_type='analytics and reporting'; no individual SQL execution or semantic review.	en	none	SELECT players.player_name, MAX(players_scores.score) as high_score, tournaments.tournament_name FROM historical_tournaments JOIN players ON historical_tournaments.player_id = players.player_id JOIN players_scores ON historical_tournaments.player_id = players_scores.player_id JOIN tournaments ON players_scores.tourname...	{"domain":"gaming industry","domain_description":"Player analytics, game performance metrics, eSports data, and virtual reality trends.","id":3074,"sql_complexity":"multiple_joins","sql_complexity_description":"two or more joins (specify inner, outer, cross)","sql_explanation":"This query finds the top 5 players with t...	2847fe00940dfc48b948f6d24368fd0f8aa2ea8e717c268cb05d998fb68e1512	gretel_sql:test:3074	multiple_joins	intermediate	Estimated from source sql_complexity='multiple_joins'; query feature labels are only proxies for difficulty, not measured solver performance. No empirical solver calibration was performed.	heuristic	needs_difficulty_calibration	not_measured	analysis_only_unvalidated	primary	task_annotation	3,073	[ "difficulty_not_calibrated", "not_individually_reviewed", "reference_not_independently_verified", "reference_or_constraint_only_no_rubric", "response_diversity_not_validated", "source_evaluation_split", "sql_dialect_and_result_equivalence_need_review", "sql_not_executed", "synthetic_reference" ]	false	true	false	task:b0f0e994af3b7aa8346219cc6b4745e38c69c75b1b3a845b9c7bac724ef85718	1	1	[ "gretel_sql" ]	held_out	curated_candidate	[]	core_candidate	[ "source_sql_complexity:multiple_joins", "sql_structure:selects=1;joins=3;statements=1;advanced=none", "combined_relational_operations_require_calibration" ]	structural_heuristic	Qwen3.6-27B	heuristic_not_model_measured
gretel_sql:37692f9c0d810a5b71074fa3949779af3e75ed50c205f7a87a54ac891b98ed51	task:2109f9d4192b8308e4d4d6286012b58289ae3b3e29019a734132d1a533e6fec1	cc347468a40650f1b91ac3222323f45d4b53aeca68286a2b4a1bb6ea18e1ddf5	gretel_sql	gretelai/synthetic_text_to_sql	740ab236e64503fba51be1101df7a1be83bf455d	synthetic_text_to_sql_test.snappy.parquet	3154	test	https://huggingface.co/datasets/gretelai/synthetic_text_to_sql/resolve/740ab236e64503fba51be1101df7a1be83bf455d/synthetic_text_to_sql_test.snappy.parquet	apache-2.0	Gretel Synthetic Text-to-SQL dataset creators; source card and author citation preserved in the download manifest.	proposed	Find renewable energy projects that are not located in the top 5 most populous cities in the world.	CREATE TABLE renewable_projects (project_name VARCHAR(255), city VARCHAR(255)); CREATE TABLE city_populations (city VARCHAR(255), population INT);	database_business_operations	database_business_operations.database_query_analysis	heuristic_unreviewed	Mapped from source sql_task_type='analytics and reporting'; no individual SQL execution or semantic review.	en	none	SELECT project_name FROM renewable_projects RP WHERE city NOT IN (SELECT city FROM (SELECT city, ROW_NUMBER() OVER (ORDER BY population DESC) as rank FROM city_populations) CP WHERE rank <= 5);	{"domain":"sustainable infrastructure","domain_description":"Green building data, renewable energy infrastructure projects, carbon offset initiatives, and smart city technology adoption.","id":3154,"sql_complexity":"window functions","sql_complexity_description":"window functions (e.g., ROW_NUMBER, LEAD, LAG, RANk, NTI...	c2a3400790d9d8744d65a670add9a37443c8778bb8cb1a14b192a52ba98fb84a	gretel_sql:test:3154	window functions	intermediate	Estimated from source sql_complexity='window functions'; query feature labels are only proxies for difficulty, not measured solver performance. No empirical solver calibration was performed.	heuristic	needs_difficulty_calibration	not_measured	analysis_only_unvalidated	primary	task_annotation	3,153	[ "difficulty_not_calibrated", "not_individually_reviewed", "reference_not_independently_verified", "reference_or_constraint_only_no_rubric", "response_diversity_not_validated", "source_evaluation_split", "sql_dialect_and_result_equivalence_need_review", "sql_not_executed", "synthetic_reference" ]	false	true	false	task:2109f9d4192b8308e4d4d6286012b58289ae3b3e29019a734132d1a533e6fec1	1	1	[ "gretel_sql" ]	held_out	curated_candidate	[]	core_candidate	[ "source_sql_complexity:window functions", "sql_structure:selects=3;joins=0;statements=1;advanced=OVER", "combined_relational_operations_require_calibration" ]	structural_heuristic	Qwen3.6-27B	heuristic_not_model_measured
gretel_sql:594bb22a607a4dac2dc7d687b22efa71c5bbfa425bfdd8a505da264245ee8f39	task:aa6585ce8e2432035e9d58e9738c5cfafe329157d5aea00e24885453a7994ffe	8ec6a4e8e11f44bf1cc99d0e36eac23a9b1ccb2a06e7b156ba182dfb80336417	gretel_sql	gretelai/synthetic_text_to_sql	740ab236e64503fba51be1101df7a1be83bf455d	synthetic_text_to_sql_test.snappy.parquet	3155	test	https://huggingface.co/datasets/gretelai/synthetic_text_to_sql/resolve/740ab236e64503fba51be1101df7a1be83bf455d/synthetic_text_to_sql_test.snappy.parquet	apache-2.0	Gretel Synthetic Text-to-SQL dataset creators; source card and author citation preserved in the download manifest.	proposed	Delete the record of the community health worker with the highest age in 'ON' province.	CREATE TABLE CommunityHealthWorkersCanada (WorkerID INT, Age INT, Gender VARCHAR(1), Province VARCHAR(2)); INSERT INTO CommunityHealthWorkersCanada (WorkerID, Age, Gender, Province) VALUES (1, 35, 'F', 'ON'), (2, 40, 'M', 'QC'), (3, 45, 'F', 'BC'), (4, 50, 'M', 'AB'), (5, 55, 'F', 'ON');	database_business_operations	database_business_operations.record_update_reconciliation	heuristic_unreviewed	Mapped from source sql_task_type='data manipulation'; no individual SQL execution or semantic review.	en	none	DELETE FROM CommunityHealthWorkersCanada WHERE WorkerID = (SELECT WorkerID FROM (SELECT WorkerID, ROW_NUMBER() OVER (PARTITION BY Province ORDER BY Age DESC) rn FROM CommunityHealthWorkersCanada) t WHERE t.rn = 1 AND t.Province = 'ON');	{"domain":"healthcare","domain_description":"Healthcare data on mental health parity, cultural competency, health equity metrics, and community health workers.","id":3155,"sql_complexity":"window functions","sql_complexity_description":"window functions (e.g., ROW_NUMBER, LEAD, LAG, RANk, NTILE, PERCENT_RANK, etc.) wit...	3f7d7257ba17c3ba0bfeeedae09e976c6ab07cef6044d7e456e92f955d7874eb	gretel_sql:test:3155	window functions	intermediate	Estimated from source sql_complexity='window functions'; query feature labels are only proxies for difficulty, not measured solver performance. No empirical solver calibration was performed.	heuristic	needs_difficulty_calibration	not_measured	analysis_only_unvalidated	primary	task_annotation	3,154	[ "difficulty_not_calibrated", "not_individually_reviewed", "reference_not_independently_verified", "reference_or_constraint_only_no_rubric", "response_diversity_not_validated", "source_evaluation_split", "sql_dialect_and_result_equivalence_need_review", "sql_not_executed", "synthetic_reference" ]	false	true	false	task:aa6585ce8e2432035e9d58e9738c5cfafe329157d5aea00e24885453a7994ffe	1	1	[ "gretel_sql" ]	held_out	curated_candidate	[]	core_candidate	[ "source_sql_complexity:window functions", "sql_structure:selects=2;joins=0;statements=1;advanced=OVER", "combined_relational_operations_require_calibration" ]	structural_heuristic	Qwen3.6-27B	heuristic_not_model_measured
gretel_sql:2fef568d981de84194069d6daceafc8de50040347d6dc015c5f2e443eace2101	task:3f54f4addb2220956a936b277dd4af7ac585a369660d81e0fefe5486c1ab3da9	25e4245afad6643f9741d3ebe03d99e92e5d2e65a0cf8802639016d2d084a7a1	gretel_sql	gretelai/synthetic_text_to_sql	740ab236e64503fba51be1101df7a1be83bf455d	synthetic_text_to_sql_test.snappy.parquet	3227	test	https://huggingface.co/datasets/gretelai/synthetic_text_to_sql/resolve/740ab236e64503fba51be1101df7a1be83bf455d/synthetic_text_to_sql_test.snappy.parquet	apache-2.0	Gretel Synthetic Text-to-SQL dataset creators; source card and author citation preserved in the download manifest.	proposed	Which onshore wells in the Permian Basin have the highest production rate?	CREATE TABLE wells (well_id INT, well_name VARCHAR(255), well_type VARCHAR(255), location VARCHAR(255)); INSERT INTO wells VALUES (1, 'Well A', 'Offshore', 'Gulf of Mexico'); INSERT INTO wells VALUES (2, 'Well B', 'Onshore', 'Permian Basin');	database_business_operations	database_business_operations.database_query_analysis	heuristic_unreviewed	Mapped from source sql_task_type='analytics and reporting'; no individual SQL execution or semantic review.	en	none	SELECT well_id, well_name, production_rate FROM (SELECT well_id, well_name, production_rate, row_number() OVER (PARTITION BY well_type ORDER BY production_rate DESC) as rn FROM well_production WHERE well_type = 'Onshore' AND location = 'Permian Basin') WHERE rn = 1;	{"domain":"oil and gas","domain_description":"Exploration data, production figures, infrastructure development, and market trends.","id":3227,"sql_complexity":"window functions","sql_complexity_description":"window functions (e.g., ROW_NUMBER, LEAD, LAG, RANk, NTILE, PERCENT_RANK, etc.) with partitioning and ordering",...	15baf98fa2c13c860019e2501b9f18d6a9e17ca9025594e0fc3917b9af5c18ea	gretel_sql:test:3227	window functions	intermediate	Estimated from source sql_complexity='window functions'; query feature labels are only proxies for difficulty, not measured solver performance. No empirical solver calibration was performed.	heuristic	needs_difficulty_calibration	not_measured	analysis_only_unvalidated	primary	task_annotation	3,226	[ "difficulty_not_calibrated", "not_individually_reviewed", "reference_not_independently_verified", "reference_or_constraint_only_no_rubric", "response_diversity_not_validated", "source_evaluation_split", "sql_dialect_and_result_equivalence_need_review", "sql_not_executed", "synthetic_reference" ]	false	true	false	task:3f54f4addb2220956a936b277dd4af7ac585a369660d81e0fefe5486c1ab3da9	1	1	[ "gretel_sql" ]	held_out	curated_candidate	[]	core_candidate	[ "source_sql_complexity:window functions", "sql_structure:selects=2;joins=0;statements=1;advanced=OVER", "combined_relational_operations_require_calibration" ]	structural_heuristic	Qwen3.6-27B	heuristic_not_model_measured
gretel_sql:685d641e285a3ac07ad6b30af20afec0c52628758a84a8f2d67d18df97bc011b	task:18659db5cf70941c27749d9fbead77d9ae2f7f6fc4402e812432419d52934ee8	fd104076f37179af216b58e724e15edac2fabe9172205cc8fe10461029b13ae7	gretel_sql	gretelai/synthetic_text_to_sql	740ab236e64503fba51be1101df7a1be83bf455d	synthetic_text_to_sql_test.snappy.parquet	3230	test	https://huggingface.co/datasets/gretelai/synthetic_text_to_sql/resolve/740ab236e64503fba51be1101df7a1be83bf455d/synthetic_text_to_sql_test.snappy.parquet	apache-2.0	Gretel Synthetic Text-to-SQL dataset creators; source card and author citation preserved in the download manifest.	proposed	Calculate the number of schools and hospitals in each community by joining the schools, hospitals, and communities tables.	CREATE TABLE communities (id INT, name VARCHAR(255)); CREATE TABLE schools (id INT, community_id INT, name VARCHAR(255)); CREATE TABLE hospitals (id INT, community_id INT, name VARCHAR(255));	database_business_operations	database_business_operations.database_query_analysis	heuristic_unreviewed	Mapped from source sql_task_type='analytics and reporting'; no individual SQL execution or semantic review.	en	none	SELECT c.name, COUNT(s.id) as school_count, COUNT(h.id) as hospital_count FROM communities c INNER JOIN schools s ON c.id = s.community_id INNER JOIN hospitals h ON c.id = h.community_id GROUP BY c.id;	{"domain":"humanitarian aid","domain_description":"Extensive data on disaster response, refugee support, community development, and advocacy in humanitarian aid.","id":3230,"sql_complexity":"multiple_joins","sql_complexity_description":"two or more joins (specify inner, outer, cross)","sql_explanation":"This query join...	284a2cd554ab3717130b782f4e122ddf4a15618497f484b8d5393a3fbe5d8f9b	gretel_sql:test:3230	multiple_joins	intermediate	Estimated from source sql_complexity='multiple_joins'; query feature labels are only proxies for difficulty, not measured solver performance. No empirical solver calibration was performed.	heuristic	needs_difficulty_calibration	not_measured	analysis_only_unvalidated	primary	task_annotation	3,229	[ "difficulty_not_calibrated", "not_individually_reviewed", "reference_not_independently_verified", "reference_or_constraint_only_no_rubric", "response_diversity_not_validated", "source_evaluation_split", "sql_dialect_and_result_equivalence_need_review", "sql_not_executed", "synthetic_reference" ]	false	true	false	task:18659db5cf70941c27749d9fbead77d9ae2f7f6fc4402e812432419d52934ee8	1	1	[ "gretel_sql" ]	held_out	curated_candidate	[]	core_candidate	[ "source_sql_complexity:multiple_joins", "sql_structure:selects=1;joins=2;statements=1;advanced=none", "combined_relational_operations_require_calibration" ]	structural_heuristic	Qwen3.6-27B	heuristic_not_model_measured
gretel_sql:dca446f0bd8118a65468d4f6dde43de9abb62a885307e338a1f25dab6c511208	task:fc68735a47ed70c1aef5a9c5e9c5b96a107ede783c74afd48e353b37e0552b86	e2f0e34c9a0d0a585ff4f87ea72273f80642f1b16a798a6dd1feeee6a2105ec3	gretel_sql	gretelai/synthetic_text_to_sql	740ab236e64503fba51be1101df7a1be83bf455d	synthetic_text_to_sql_test.snappy.parquet	3273	test	https://huggingface.co/datasets/gretelai/synthetic_text_to_sql/resolve/740ab236e64503fba51be1101df7a1be83bf455d/synthetic_text_to_sql_test.snappy.parquet	apache-2.0	Gretel Synthetic Text-to-SQL dataset creators; source card and author citation preserved in the download manifest.	proposed	List all departments, the number of chemicals they handle, and the number of machines used in their manufacturing processes.	CREATE TABLE Departments (DepartmentID INT, DepartmentName VARCHAR(20)); INSERT INTO Departments (DepartmentID, DepartmentName) VALUES (1, 'Manufacturing'), (2, 'Engineering'), (3, 'Chemical Safety'), (4, 'Product Development'); CREATE TABLE Chemicals (ChemicalID INT, ChemicalName VARCHAR(20), DepartmentID INT); INSERT...	database_business_operations	database_business_operations.database_query_analysis	heuristic_unreviewed	Mapped from source sql_task_type='analytics and reporting'; no individual SQL execution or semantic review.	en	none	SELECT D.DepartmentName, COUNT(C.ChemicalID) AS ChemicalCount, COUNT(M.MachineID) AS MachineCount FROM Departments D LEFT JOIN Chemicals C ON D.DepartmentID = C.DepartmentID LEFT JOIN Machinery M ON D.DepartmentID = M.DepartmentID GROUP BY D.DepartmentName;	{"domain":"chemicals","domain_description":"Chemical manufacturing data, safety protocols, environmental impact assessments, and product innovation metrics.","id":3273,"sql_complexity":"multiple_joins","sql_complexity_description":"two or more joins (specify inner, outer, cross)","sql_explanation":"We select the Depart...	c7ea4f22017d93a95e578c9d9b972a19070612d958ad67fda58b92e5387b533d	gretel_sql:test:3273	multiple_joins	intermediate	Estimated from source sql_complexity='multiple_joins'; query feature labels are only proxies for difficulty, not measured solver performance. No empirical solver calibration was performed.	heuristic	needs_difficulty_calibration	not_measured	analysis_only_unvalidated	primary	task_annotation	3,272	[ "difficulty_not_calibrated", "not_individually_reviewed", "reference_not_independently_verified", "reference_or_constraint_only_no_rubric", "response_diversity_not_validated", "source_evaluation_split", "sql_dialect_and_result_equivalence_need_review", "sql_not_executed", "synthetic_reference" ]	false	true	false	task:fc68735a47ed70c1aef5a9c5e9c5b96a107ede783c74afd48e353b37e0552b86	1	1	[ "gretel_sql" ]	held_out	curated_candidate	[]	core_candidate	[ "source_sql_complexity:multiple_joins", "sql_structure:selects=1;joins=2;statements=1;advanced=none", "combined_relational_operations_require_calibration" ]	structural_heuristic	Qwen3.6-27B	heuristic_not_model_measured
gretel_sql:d29ca44f0863756af9bb2e0bd26737d733b14775e182679dca8e6aa12bb1f921	task:5e0a0ee30b96348836675532bba41e26320aa226711e8336999b62ecdd63d730	0eaa8901de47b94031b59b30dc5633f8be795d64f11288499684e1e6e9698bc5	gretel_sql	gretelai/synthetic_text_to_sql	740ab236e64503fba51be1101df7a1be83bf455d	synthetic_text_to_sql_test.snappy.parquet	3334	test	https://huggingface.co/datasets/gretelai/synthetic_text_to_sql/resolve/740ab236e64503fba51be1101df7a1be83bf455d/synthetic_text_to_sql_test.snappy.parquet	apache-2.0	Gretel Synthetic Text-to-SQL dataset creators; source card and author citation preserved in the download manifest.	proposed	What is the name of the vendor with the highest revenue from organic products?	CREATE TABLE Vendors (VendorID INT, VendorName TEXT, Country TEXT);CREATE TABLE Products (ProductID INT, ProductName TEXT, Price DECIMAL, Organic BOOLEAN, VendorID INT); INSERT INTO Vendors VALUES (1, 'VendorF', 'UK'), (2, 'VendorG', 'UK'); INSERT INTO Products VALUES (1, 'Carrot', 0.6, true, 1), (2, 'Broccoli', 1.2, t...	database_business_operations	database_business_operations.database_query_analysis	heuristic_unreviewed	Mapped from source sql_task_type='analytics and reporting'; no individual SQL execution or semantic review.	en	none	SELECT v.VendorName FROM Vendors v JOIN Products p ON v.VendorID = p.VendorID WHERE p.Organic = true GROUP BY v.VendorID, v.VendorName HAVING SUM(Price) = (SELECT MAX(TotalPrice) FROM (SELECT SUM(Price) AS TotalPrice FROM Products p JOIN Vendors v ON p.VendorID = v.VendorID WHERE p.Organic = true GROUP BY v.VendorID) t...	{"domain":"retail","domain_description":"Retail data on circular supply chains, ethical labor practices, product transparency, and consumer education.","id":3334,"sql_complexity":"multiple_joins","sql_complexity_description":"two or more joins (specify inner, outer, cross)","sql_explanation":"This query finds the name ...	e4da024f97241750b46dc8d1c5d70f9f71f59c7d36390fb8236a8023d6a11c54	gretel_sql:test:3334	multiple_joins	intermediate	Estimated from source sql_complexity='multiple_joins'; query feature labels are only proxies for difficulty, not measured solver performance. No empirical solver calibration was performed.	heuristic	needs_difficulty_calibration	not_measured	analysis_only_unvalidated	primary	task_annotation	3,333	[ "difficulty_not_calibrated", "not_individually_reviewed", "reference_not_independently_verified", "reference_or_constraint_only_no_rubric", "response_diversity_not_validated", "source_evaluation_split", "sql_dialect_and_result_equivalence_need_review", "sql_not_executed", "synthetic_reference" ]	false	true	false	task:5e0a0ee30b96348836675532bba41e26320aa226711e8336999b62ecdd63d730	1	1	[ "gretel_sql" ]	held_out	curated_candidate	[]	core_candidate	[ "source_sql_complexity:multiple_joins", "sql_structure:selects=3;joins=2;statements=1;advanced=none", "combined_relational_operations_require_calibration" ]	structural_heuristic	Qwen3.6-27B	heuristic_not_model_measured
gretel_sql:06aa6f20616125ec535b79ffdd6712b71e656c64f2576154825843c43785f233	task:56d7a2732424b4c48b9f01ab9af22c245671acd7868697899ed25f3315f48a8d	47850b8e9413e2a89df3ae82481228de01505e8bc55f8866a8d6c288f972474a	gretel_sql	gretelai/synthetic_text_to_sql	740ab236e64503fba51be1101df7a1be83bf455d	synthetic_text_to_sql_test.snappy.parquet	3380	test	https://huggingface.co/datasets/gretelai/synthetic_text_to_sql/resolve/740ab236e64503fba51be1101df7a1be83bf455d/synthetic_text_to_sql_test.snappy.parquet	apache-2.0	Gretel Synthetic Text-to-SQL dataset creators; source card and author citation preserved in the download manifest.	proposed	List all smart contracts and their respective total transaction volume for Q3 2021, ordered by the smart contracts with the highest transaction volume in Q3 2021.	CREATE TABLE SmartContracts (contract_id INT, contract_name VARCHAR(255), transaction_volume DECIMAL(18,2)); INSERT INTO SmartContracts (contract_id, contract_name, transaction_volume) VALUES (1, 'Uniswap', 156789.12), (2, 'SushiSwap', 256789.12), (3, 'Aave', 356789.12), (4, 'Compound', 456789.12), (5, 'Yearn Finance',...	database_business_operations	database_business_operations.database_query_analysis	heuristic_unreviewed	Mapped from source sql_task_type='analytics and reporting'; no individual SQL execution or semantic review.	en	none	SELECT contract_name, transaction_volume FROM (SELECT contract_name, transaction_volume, RANK() OVER (ORDER BY transaction_volume DESC) as rank FROM SmartContracts WHERE SmartContracts.transaction_date BETWEEN '2021-07-01' AND '2021-09-30') AS ranked_contracts ORDER BY rank;	{"domain":"blockchain","domain_description":"Comprehensive data on smart contracts, decentralized applications, digital assets, and regulatory frameworks in blockchain.","id":3380,"sql_complexity":"window functions","sql_complexity_description":"window functions (e.g., ROW_NUMBER, LEAD, LAG, RANk, NTILE, PERCENT_RANK, ...	6dbc1a2c19513be6a8d40c7c076facaf5dfca77a654eae76a6d7ead2e0eb2e7f	gretel_sql:test:3380	window functions	intermediate	Estimated from source sql_complexity='window functions'; query feature labels are only proxies for difficulty, not measured solver performance. No empirical solver calibration was performed.	heuristic	needs_difficulty_calibration	not_measured	analysis_only_unvalidated	primary	task_annotation	3,379	[ "difficulty_not_calibrated", "not_individually_reviewed", "reference_not_independently_verified", "reference_or_constraint_only_no_rubric", "response_diversity_not_validated", "source_evaluation_split", "sql_dialect_and_result_equivalence_need_review", "sql_not_executed", "synthetic_reference" ]	false	true	false	task:56d7a2732424b4c48b9f01ab9af22c245671acd7868697899ed25f3315f48a8d	1	1	[ "gretel_sql" ]	held_out	curated_candidate	[]	core_candidate	[ "source_sql_complexity:window functions", "sql_structure:selects=2;joins=0;statements=1;advanced=OVER", "combined_relational_operations_require_calibration" ]	structural_heuristic	Qwen3.6-27B	heuristic_not_model_measured
gretel_sql:a951a9dbc6b8cc46c34519127dec6fa48b551386255287dd83fb5a11b67a85bb	task:74138b9319a7f55e58d588de5583d61813cec5fca05269dec86dbdd305e2fd8f	ca8214dfc8ffac268cebf991859c45dab7de2f38899cd59647632591215da5a5	gretel_sql	gretelai/synthetic_text_to_sql	740ab236e64503fba51be1101df7a1be83bf455d	synthetic_text_to_sql_test.snappy.parquet	3394	test	https://huggingface.co/datasets/gretelai/synthetic_text_to_sql/resolve/740ab236e64503fba51be1101df7a1be83bf455d/synthetic_text_to_sql_test.snappy.parquet	apache-2.0	Gretel Synthetic Text-to-SQL dataset creators; source card and author citation preserved in the download manifest.	proposed	List all the projects implemented in Nepal, their implementing organizations, and the total budget for each project, sorted by the implementing organizations.	CREATE TABLE projects (id INT, name TEXT, country TEXT, budget INT); INSERT INTO projects VALUES (1, 'School Rebuilding', 'Nepal', 100000); CREATE TABLE organizations_projects (organization_id INT, project_id INT); INSERT INTO organizations_projects VALUES (1, 1); CREATE TABLE organizations (id INT, name TEXT); INSERT ...	database_business_operations	database_business_operations.database_query_analysis	heuristic_unreviewed	Mapped from source sql_task_type='analytics and reporting'; no individual SQL execution or semantic review.	en	none	SELECT o.name, p.name, SUM(p.budget) FROM projects p INNER JOIN organizations_projects op ON p.id = op.project_id INNER JOIN organizations o ON op.organization_id = o.id WHERE p.country = 'Nepal' GROUP BY o.id, p.name ORDER BY o.name;	{"domain":"humanitarian aid","domain_description":"Extensive data on disaster response, refugee support, community development, and advocacy in humanitarian aid.","id":3394,"sql_complexity":"multiple_joins","sql_complexity_description":"two or more joins (specify inner, outer, cross)","sql_explanation":"The SQL query f...	2c71a0169c2c021bcbe65ffebb1ae15a17f9ecd66a153551a89377df97575b3d	gretel_sql:test:3394	multiple_joins	intermediate	Estimated from source sql_complexity='multiple_joins'; query feature labels are only proxies for difficulty, not measured solver performance. No empirical solver calibration was performed.	heuristic	needs_difficulty_calibration	not_measured	analysis_only_unvalidated	primary	task_annotation	3,393	[ "difficulty_not_calibrated", "not_individually_reviewed", "reference_not_independently_verified", "reference_or_constraint_only_no_rubric", "response_diversity_not_validated", "source_evaluation_split", "sql_dialect_and_result_equivalence_need_review", "sql_not_executed", "synthetic_reference" ]	false	true	false	task:74138b9319a7f55e58d588de5583d61813cec5fca05269dec86dbdd305e2fd8f	1	1	[ "gretel_sql" ]	held_out	curated_candidate	[]	core_candidate	[ "source_sql_complexity:multiple_joins", "sql_structure:selects=1;joins=2;statements=1;advanced=none", "combined_relational_operations_require_calibration" ]	structural_heuristic	Qwen3.6-27B	heuristic_not_model_measured
gretel_sql:b53ecb21605ceef3e57cd2692b2374ab7f028ece9defa3934a3ba1b5ea09902c	task:1a8eb895aa875fc37881d89e81ad0390a9cdcaa5d6ea832095a94ed0aec26d36	e3215462eedb1036dba7f4dd4bd9905dda440ef1a6b86d4bfbccc48771fc08e1	gretel_sql	gretelai/synthetic_text_to_sql	740ab236e64503fba51be1101df7a1be83bf455d	synthetic_text_to_sql_test.snappy.parquet	3411	test	https://huggingface.co/datasets/gretelai/synthetic_text_to_sql/resolve/740ab236e64503fba51be1101df7a1be83bf455d/synthetic_text_to_sql_test.snappy.parquet	apache-2.0	Gretel Synthetic Text-to-SQL dataset creators; source card and author citation preserved in the download manifest.	proposed	Which countries have the highest average spending on virtual tours per user?	CREATE TABLE Users (UserID INT, CountryID INT); INSERT INTO Users (UserID, CountryID) VALUES (1, 1), (2, 1), (3, 2), (4, 2), (5, 3), (6, 3), (7, 4), (8, 4); CREATE TABLE Spending (SpendingID INT, UserID INT, VirtualTourID INT, Amount DECIMAL(10,2)); INSERT INTO Spending (SpendingID, UserID, VirtualTourID, Amount) VALUE...	database_business_operations	database_business_operations.database_query_analysis	heuristic_unreviewed	Mapped from source sql_task_type='analytics and reporting'; no individual SQL execution or semantic review.	en	none	SELECT U.CountryName, AVG(S.Amount) as AvgSpending FROM Users U INNER JOIN Spending S ON U.UserID = S.UserID INNER JOIN VirtualTours VT ON S.VirtualTourID = VT.VirtualTourID GROUP BY U.CountryID ORDER BY AvgSpending DESC;	{"domain":"hospitality","domain_description":"Hospitality data on sustainable tourism, cultural heritage preservation, virtual tourism, and local economic impact.","id":3411,"sql_complexity":"multiple_joins","sql_complexity_description":"two or more joins (specify inner, outer, cross)","sql_explanation":"This SQL query...	f77e4e0317f0a1847a89755ee3e61b6db6313641a6b39e3806f004899a11e81e	gretel_sql:test:3411	multiple_joins	intermediate	Estimated from source sql_complexity='multiple_joins'; query feature labels are only proxies for difficulty, not measured solver performance. No empirical solver calibration was performed.	heuristic	needs_difficulty_calibration	not_measured	analysis_only_unvalidated	primary	task_annotation	3,410	[ "difficulty_not_calibrated", "not_individually_reviewed", "reference_not_independently_verified", "reference_or_constraint_only_no_rubric", "response_diversity_not_validated", "source_evaluation_split", "sql_dialect_and_result_equivalence_need_review", "sql_not_executed", "synthetic_reference" ]	false	true	false	task:1a8eb895aa875fc37881d89e81ad0390a9cdcaa5d6ea832095a94ed0aec26d36	1	1	[ "gretel_sql" ]	held_out	curated_candidate	[]	core_candidate	[ "source_sql_complexity:multiple_joins", "sql_structure:selects=1;joins=2;statements=1;advanced=none", "combined_relational_operations_require_calibration" ]	structural_heuristic	Qwen3.6-27B	heuristic_not_model_measured
gretel_sql:e3aa74e1c2f6428ecfcbd9b28c37b8efe553b7f5c68587b2a0f0ca2649e07ecf	task:c58a226ebd54ab034c5e1cab29a5f8bc810cbd081004ea4151f98c07ddc52ef7	03913aea434c84a20173b5adfc69fcba42c5c7ef15e41be4fb722a0813f799d0	gretel_sql	gretelai/synthetic_text_to_sql	740ab236e64503fba51be1101df7a1be83bf455d	synthetic_text_to_sql_test.snappy.parquet	3439	test	https://huggingface.co/datasets/gretelai/synthetic_text_to_sql/resolve/740ab236e64503fba51be1101df7a1be83bf455d/synthetic_text_to_sql_test.snappy.parquet	apache-2.0	Gretel Synthetic Text-to-SQL dataset creators; source card and author citation preserved in the download manifest.	proposed	How many menu items contain ingredients sourced from local suppliers for each restaurant?	CREATE TABLE restaurants (id INT, name VARCHAR(50), location VARCHAR(50)); INSERT INTO restaurants VALUES (1, 'Restaurant A', 'City A'); INSERT INTO restaurants VALUES (2, 'Restaurant B', 'City B'); CREATE TABLE menu_items (id INT, name VARCHAR(50), restaurant_id INT, price DECIMAL(5,2)); INSERT INTO menu_items VALUES ...	database_business_operations	database_business_operations.database_query_analysis	heuristic_unreviewed	Mapped from source sql_task_type='analytics and reporting'; no individual SQL execution or semantic review.	en	none	SELECT r.name, COUNT(DISTINCT mi.id) as num_local_items FROM restaurants r JOIN menu_items mi ON r.id = mi.restaurant_id JOIN ingredients i ON mi.id = i.menu_item_id WHERE i.local_source = TRUE GROUP BY r.name;	{"domain":"restaurant operations","domain_description":"Restaurant revenue data, menu engineering information, food safety inspection records, and sustainable sourcing practices.","id":3439,"sql_complexity":"multiple_joins","sql_complexity_description":"two or more joins (specify inner, outer, cross)","sql_explanation"...	8581126afce6fd4338dbfdf9aa721809450cd1060f25d2c827c9c01e3b8e2a41	gretel_sql:test:3439	multiple_joins	intermediate	Estimated from source sql_complexity='multiple_joins'; query feature labels are only proxies for difficulty, not measured solver performance. No empirical solver calibration was performed.	heuristic	needs_difficulty_calibration	not_measured	analysis_only_unvalidated	primary	task_annotation	3,438	[ "difficulty_not_calibrated", "not_individually_reviewed", "reference_not_independently_verified", "reference_or_constraint_only_no_rubric", "response_diversity_not_validated", "source_evaluation_split", "sql_dialect_and_result_equivalence_need_review", "sql_not_executed", "synthetic_reference" ]	false	true	false	task:c58a226ebd54ab034c5e1cab29a5f8bc810cbd081004ea4151f98c07ddc52ef7	1	1	[ "gretel_sql" ]	held_out	curated_candidate	[]	core_candidate	[ "source_sql_complexity:multiple_joins", "sql_structure:selects=1;joins=2;statements=1;advanced=none", "combined_relational_operations_require_calibration" ]	structural_heuristic	Qwen3.6-27B	heuristic_not_model_measured
gretel_sql:a923b282902aa292117ed2e138461b5a47ef45f38d0636ba50853ac462f408e5	task:9500cfc0c931d716a456689ebcf6f5bd3bcda57fa5d16feaeb03019f21173403	ceb2fe961e2d9e211c99822e74a30bd9da50a1234a44621925ee6cdcdf398f74	gretel_sql	gretelai/synthetic_text_to_sql	740ab236e64503fba51be1101df7a1be83bf455d	synthetic_text_to_sql_test.snappy.parquet	3464	test	https://huggingface.co/datasets/gretelai/synthetic_text_to_sql/resolve/740ab236e64503fba51be1101df7a1be83bf455d/synthetic_text_to_sql_test.snappy.parquet	apache-2.0	Gretel Synthetic Text-to-SQL dataset creators; source card and author citation preserved in the download manifest.	proposed	Identify the chemical plants in Canada with safety violation costs higher than their preceding plant.	CREATE TABLE chemical_plants (plant_id INT, plant_name VARCHAR(50), country VARCHAR(50), safety_violation_cost DECIMAL(10,2), plant_order INT); INSERT INTO chemical_plants (plant_id, plant_name, country, safety_violation_cost, plant_order) VALUES (1, 'Plant A', 'Canada', 5000, 1), (2, 'Plant B', 'Canada', 7000, 2), (3,...	database_business_operations	database_business_operations.database_query_analysis	heuristic_unreviewed	Mapped from source sql_task_type='analytics and reporting'; no individual SQL execution or semantic review.	en	none	SELECT plant_id, plant_name, safety_violation_cost FROM (SELECT plant_id, plant_name, safety_violation_cost, LAG(safety_violation_cost) OVER (PARTITION BY country ORDER BY plant_order) AS lag_value FROM chemical_plants WHERE country = 'Canada') tmp WHERE safety_violation_cost > lag_value;	{"domain":"chemicals","domain_description":"Chemical manufacturing data, safety protocols, environmental impact assessments, and product innovation metrics.","id":3464,"sql_complexity":"window functions","sql_complexity_description":"window functions (e.g., ROW_NUMBER, LEAD, LAG, RANk, NTILE, PERCENT_RANK, etc.) with p...	061685d3658143d143679b599598a08b49344797cb553653786042836d30c8b4	gretel_sql:test:3464	window functions	intermediate	Estimated from source sql_complexity='window functions'; query feature labels are only proxies for difficulty, not measured solver performance. No empirical solver calibration was performed.	heuristic	needs_difficulty_calibration	not_measured	analysis_only_unvalidated	primary	task_annotation	3,463	[ "difficulty_not_calibrated", "not_individually_reviewed", "reference_not_independently_verified", "reference_or_constraint_only_no_rubric", "response_diversity_not_validated", "source_evaluation_split", "sql_dialect_and_result_equivalence_need_review", "sql_not_executed", "synthetic_reference" ]	false	true	false	task:9500cfc0c931d716a456689ebcf6f5bd3bcda57fa5d16feaeb03019f21173403	1	1	[ "gretel_sql" ]	held_out	curated_candidate	[]	core_candidate	[ "source_sql_complexity:window functions", "sql_structure:selects=2;joins=0;statements=1;advanced=OVER", "combined_relational_operations_require_calibration" ]	structural_heuristic	Qwen3.6-27B	heuristic_not_model_measured
gretel_sql:552bf2c3851ed99608a9e6959065eb05267a87d40aa1fe13008a4a25393663e1	task:dbecb377b4fe3e2d4618fae561f9c270b3ae23010fed65c02d80c9e9497db54f	37063ea4e20629590e082832750b2e4eb70a25d9992e945db87351e892870b46	gretel_sql	gretelai/synthetic_text_to_sql	740ab236e64503fba51be1101df7a1be83bf455d	synthetic_text_to_sql_test.snappy.parquet	3475	test	https://huggingface.co/datasets/gretelai/synthetic_text_to_sql/resolve/740ab236e64503fba51be1101df7a1be83bf455d/synthetic_text_to_sql_test.snappy.parquet	apache-2.0	Gretel Synthetic Text-to-SQL dataset creators; source card and author citation preserved in the download manifest.	proposed	What is the density of trees (trees per hectare) by tree type in each country?	CREATE TABLE countries (id INT, name VARCHAR(255)); INSERT INTO countries (id, name) VALUES (1, 'Canada'), (2, 'USA'); CREATE TABLE tree_densities (id INT, country_id INT, tree_type_id INT, trees_per_hectare INT); INSERT INTO tree_densities (id, country_id, tree_type_id, trees_per_hectare) VALUES (1, 1, 1, 100), (2, 1,...	database_business_operations	database_business_operations.database_query_analysis	heuristic_unreviewed	Mapped from source sql_task_type='analytics and reporting'; no individual SQL execution or semantic review.	en	none	SELECT c.name, tt.name, AVG(td.trees_per_hectare) avg_trees_per_hectare FROM tree_densities td JOIN countries c ON td.country_id = c.id JOIN tree_types tt ON td.tree_type_id = tt.id GROUP BY c.name, tt.name;	{"domain":"forestry","domain_description":"Comprehensive data on sustainable forest management, timber production, wildlife habitat, and carbon sequestration in forestry.","id":3475,"sql_complexity":"multiple_joins","sql_complexity_description":"two or more joins (specify inner, outer, cross)","sql_explanation":"This q...	6e67bed09fa125876b3a1cbdba6e993bc544c9fb6e27774eebe69cfdc6b4b321	gretel_sql:test:3475	multiple_joins	intermediate	Estimated from source sql_complexity='multiple_joins'; query feature labels are only proxies for difficulty, not measured solver performance. No empirical solver calibration was performed.	heuristic	needs_difficulty_calibration	not_measured	analysis_only_unvalidated	primary	task_annotation	3,474	[ "difficulty_not_calibrated", "not_individually_reviewed", "reference_not_independently_verified", "reference_or_constraint_only_no_rubric", "response_diversity_not_validated", "source_evaluation_split", "sql_dialect_and_result_equivalence_need_review", "sql_not_executed", "synthetic_reference" ]	false	true	false	task:dbecb377b4fe3e2d4618fae561f9c270b3ae23010fed65c02d80c9e9497db54f	1	1	[ "gretel_sql" ]	held_out	curated_candidate	[]	core_candidate	[ "source_sql_complexity:multiple_joins", "sql_structure:selects=1;joins=2;statements=1;advanced=none", "combined_relational_operations_require_calibration" ]	structural_heuristic	Qwen3.6-27B	heuristic_not_model_measured
gretel_sql:b86313c1bd447f3b6055fc46f452323aebeaa575042e6670b8247cfeae69651c	task:0971a63dc7ff90908bd44a7892f9a72ed0a86f91760af09509b9d39056490ee2	f8089a0218ae11eb6fa0a2e7c3f9db83ed30a93720210bb16941258f30b867fd	gretel_sql	gretelai/synthetic_text_to_sql	740ab236e64503fba51be1101df7a1be83bf455d	synthetic_text_to_sql_test.snappy.parquet	3491	test	https://huggingface.co/datasets/gretelai/synthetic_text_to_sql/resolve/740ab236e64503fba51be1101df7a1be83bf455d/synthetic_text_to_sql_test.snappy.parquet	apache-2.0	Gretel Synthetic Text-to-SQL dataset creators; source card and author citation preserved in the download manifest.	proposed	What is the percentage of users in Canada who have used the hashtag #travel?	CREATE TABLE posts (id INT, user_id INT, hashtags TEXT); INSERT INTO posts (id, user_id, hashtags) VALUES (1, 1, '#travel'), (2, 1, '#food'), (3, 2, '#travel'), (4, 3, '#art'), (5, 4, '#music'); CREATE TABLE users (id INT, country VARCHAR(2)); INSERT INTO users (id, country) VALUES (1, 'CA'), (2, 'US'), (3, 'CA'), (4, ...	database_business_operations	database_business_operations.database_query_analysis	heuristic_unreviewed	Mapped from source sql_task_type='analytics and reporting'; no individual SQL execution or semantic review.	en	none	SELECT 100.0 * COUNT(DISTINCT user_id) / (SELECT COUNT(DISTINCT user_id) FROM posts JOIN users ON posts.user_id = users.id WHERE users.country = 'CA') as percentage_travel FROM posts JOIN users ON posts.user_id = users.id WHERE hashtags LIKE '%#travel%' AND users.country = 'CA';	{"domain":"social media","domain_description":"Extensive data on user behavior, content trends, advertising performance, and data privacy in social media.","id":3491,"sql_complexity":"multiple_joins","sql_complexity_description":"two or more joins (specify inner, outer, cross)","sql_explanation":"This query calculates ...	71f112b1aa3da847b4bc24cb0aa953cb08e19cb8c5f26259a5b9fd1085c95dd1	gretel_sql:test:3491	multiple_joins	intermediate	Estimated from source sql_complexity='multiple_joins'; query feature labels are only proxies for difficulty, not measured solver performance. No empirical solver calibration was performed.	heuristic	needs_difficulty_calibration	not_measured	analysis_only_unvalidated	primary	task_annotation	3,490	[ "difficulty_not_calibrated", "not_individually_reviewed", "reference_not_independently_verified", "reference_or_constraint_only_no_rubric", "response_diversity_not_validated", "source_evaluation_split", "sql_dialect_and_result_equivalence_need_review", "sql_not_executed", "synthetic_reference" ]	false	true	false	task:0971a63dc7ff90908bd44a7892f9a72ed0a86f91760af09509b9d39056490ee2	1	1	[ "gretel_sql" ]	held_out	curated_candidate	[]	core_candidate	[ "source_sql_complexity:multiple_joins", "sql_structure:selects=2;joins=2;statements=1;advanced=none", "combined_relational_operations_require_calibration" ]	structural_heuristic	Qwen3.6-27B	heuristic_not_model_measured
gretel_sql:3e2f2ded2c48ee81c7256ad0aa71909cd4a9148fda5b461a8d1388a2ceeff6da	task:6791561224282abe028ab7e14e239375901d45849a62ca18c9903f1df524f516	7aaff944845e50046a1af1ee57db8452ea925be5d781f369e43002a079462cf9	gretel_sql	gretelai/synthetic_text_to_sql	740ab236e64503fba51be1101df7a1be83bf455d	synthetic_text_to_sql_test.snappy.parquet	3544	test	https://huggingface.co/datasets/gretelai/synthetic_text_to_sql/resolve/740ab236e64503fba51be1101df7a1be83bf455d/synthetic_text_to_sql_test.snappy.parquet	apache-2.0	Gretel Synthetic Text-to-SQL dataset creators; source card and author citation preserved in the download manifest.	proposed	How many students with disabilities are enrolled in each region's universities?	CREATE TABLE Regions (RegionID INT PRIMARY KEY, RegionName VARCHAR(50)); CREATE TABLE Universities (UniversityID INT PRIMARY KEY, UniversityName VARCHAR(50), RegionID INT, FOREIGN KEY (RegionID) REFERENCES Regions(RegionID)); CREATE TABLE Students (StudentID INT PRIMARY KEY, StudentName VARCHAR(50), Disability BOOLEAN,...	database_business_operations	database_business_operations.database_query_analysis	heuristic_unreviewed	Mapped from source sql_task_type='analytics and reporting'; no individual SQL execution or semantic review.	en	none	SELECT r.RegionName, COUNT(s.StudentID) as StudentCount FROM Regions r JOIN Universities u ON r.RegionID = u.RegionID JOIN Students s ON u.UniversityID = s.UniversityID WHERE s.Disability = TRUE GROUP BY r.RegionName;	{"domain":"disability services","domain_description":"Comprehensive data on disability accommodations, support programs, policy advocacy, and inclusion efforts in disability services.","id":3544,"sql_complexity":"multiple_joins","sql_complexity_description":"two or more joins (specify inner, outer, cross)","sql_explana...	cc40e9b01b843979e10e1dc0bdbf46b6f602636cc9bc58e0f58d1f835c2f57f7	gretel_sql:test:3544	multiple_joins	intermediate	Estimated from source sql_complexity='multiple_joins'; query feature labels are only proxies for difficulty, not measured solver performance. No empirical solver calibration was performed.	heuristic	needs_difficulty_calibration	not_measured	analysis_only_unvalidated	primary	task_annotation	3,543	[ "difficulty_not_calibrated", "not_individually_reviewed", "reference_not_independently_verified", "reference_or_constraint_only_no_rubric", "response_diversity_not_validated", "source_evaluation_split", "sql_dialect_and_result_equivalence_need_review", "sql_not_executed", "synthetic_reference" ]	false	true	false	task:6791561224282abe028ab7e14e239375901d45849a62ca18c9903f1df524f516	1	1	[ "gretel_sql" ]	held_out	curated_candidate	[]	core_candidate	[ "source_sql_complexity:multiple_joins", "sql_structure:selects=1;joins=2;statements=1;advanced=none", "combined_relational_operations_require_calibration" ]	structural_heuristic	Qwen3.6-27B	heuristic_not_model_measured
gretel_sql:0cfc308595f74dec9047a411eb7197a544c61cda15b9459977c621bed5de842f	task:05143ece1e2bc1599c0459f417fe2701ccc32acabbcd30fb81d4f8d16f1b34a2	5f45511daf0f6b895f8faab47bb50226bb9e7535d9a61241533be1cb9623e2a5	gretel_sql	gretelai/synthetic_text_to_sql	740ab236e64503fba51be1101df7a1be83bf455d	synthetic_text_to_sql_test.snappy.parquet	3548	test	https://huggingface.co/datasets/gretelai/synthetic_text_to_sql/resolve/740ab236e64503fba51be1101df7a1be83bf455d/synthetic_text_to_sql_test.snappy.parquet	apache-2.0	Gretel Synthetic Text-to-SQL dataset creators; source card and author citation preserved in the download manifest.	proposed	What is the average carbon sequestration rate per tree species by region?	CREATE TABLE tree_species (species_id INT, species_name VARCHAR(100), avg_carbon_sequestration_rate DECIMAL(5,2)); INSERT INTO tree_species (species_id, species_name, avg_carbon_sequestration_rate) VALUES (1, 'Oak', 15.5), (2, 'Pine', 12.8), (3, 'Maple', 18.2), (4, 'Birch', 10.9); CREATE TABLE regions (region_id INT, r...	database_business_operations	database_business_operations.database_query_analysis	heuristic_unreviewed	Mapped from source sql_task_type='analytics and reporting'; no individual SQL execution or semantic review.	en	none	SELECT r.region_name, AVG(ts.avg_carbon_sequestration_rate) as avg_rate FROM tree_regions tr JOIN tree_species ts ON tr.species_id = ts.species_id JOIN regions r ON tr.region_id = r.region_id GROUP BY r.region_name;	{"domain":"forestry","domain_description":"Comprehensive data on sustainable forest management, timber production, wildlife habitat, and carbon sequestration in forestry.","id":3548,"sql_complexity":"multiple_joins","sql_complexity_description":"two or more joins (specify inner, outer, cross)","sql_explanation":"This q...	62c69bac67fdf86cf9155ed71ce6068a4c2e1c3156c7d8b4f2de64c77c116b05	gretel_sql:test:3548	multiple_joins	intermediate	Estimated from source sql_complexity='multiple_joins'; query feature labels are only proxies for difficulty, not measured solver performance. No empirical solver calibration was performed.	heuristic	needs_difficulty_calibration	not_measured	analysis_only_unvalidated	primary	task_annotation	3,547	[ "difficulty_not_calibrated", "not_individually_reviewed", "reference_not_independently_verified", "reference_or_constraint_only_no_rubric", "response_diversity_not_validated", "source_evaluation_split", "sql_dialect_and_result_equivalence_need_review", "sql_not_executed", "synthetic_reference" ]	false	true	false	task:05143ece1e2bc1599c0459f417fe2701ccc32acabbcd30fb81d4f8d16f1b34a2	1	1	[ "gretel_sql" ]	held_out	curated_candidate	[]	core_candidate	[ "source_sql_complexity:multiple_joins", "sql_structure:selects=1;joins=2;statements=1;advanced=none", "combined_relational_operations_require_calibration" ]	structural_heuristic	Qwen3.6-27B	heuristic_not_model_measured
gretel_sql:791ddef1b6269a71b3c4585feea8d7735f611a3cdaf9f9bd0940a85c330d6b35	task:3fa5cf1b2db6236e0e3809562c5078de6b333859bed5a63cbe29f842915ec8f4	553dea3de869494a21ff6c44790b2fedfe412945ccd2ec9a85e8f0e2a25d1b76	gretel_sql	gretelai/synthetic_text_to_sql	740ab236e64503fba51be1101df7a1be83bf455d	synthetic_text_to_sql_test.snappy.parquet	3551	test	https://huggingface.co/datasets/gretelai/synthetic_text_to_sql/resolve/740ab236e64503fba51be1101df7a1be83bf455d/synthetic_text_to_sql_test.snappy.parquet	apache-2.0	Gretel Synthetic Text-to-SQL dataset creators; source card and author citation preserved in the download manifest.	proposed	List all artists who have performed in New York and Chicago, along with their highest-earning performance.	CREATE TABLE artist_events (artist_id INT, event_id INT, earnings DECIMAL(5,2)); CREATE TABLE artists (id INT, name VARCHAR(50)); CREATE TABLE events (id INT, city VARCHAR(20));	database_business_operations	database_business_operations.database_query_analysis	heuristic_unreviewed	Mapped from source sql_task_type='analytics and reporting'; no individual SQL execution or semantic review.	en	none	SELECT artists.name, MAX(artist_events.earnings) FROM artists INNER JOIN artist_events ON artists.id = artist_events.artist_id INNER JOIN events ON artist_events.event_id = events.id WHERE city IN ('New York', 'Chicago') GROUP BY artists.name;	{"domain":"arts operations and management","domain_description":"Data on art collections, cultural event attendance, artist demographics, museum operations, and heritage preservation efforts.","id":3551,"sql_complexity":"multiple_joins","sql_complexity_description":"two or more joins (specify inner, outer, cross)","sql...	5ffbac704ee803e62ff4529ee8e9caa892ef88b3fd7ccd3434421efb47bf49e3	gretel_sql:test:3551	multiple_joins	intermediate	Estimated from source sql_complexity='multiple_joins'; query feature labels are only proxies for difficulty, not measured solver performance. No empirical solver calibration was performed.	heuristic	needs_difficulty_calibration	not_measured	analysis_only_unvalidated	primary	task_annotation	3,550	[ "difficulty_not_calibrated", "not_individually_reviewed", "reference_not_independently_verified", "reference_or_constraint_only_no_rubric", "response_diversity_not_validated", "source_evaluation_split", "sql_dialect_and_result_equivalence_need_review", "sql_not_executed", "synthetic_reference" ]	false	true	false	task:3fa5cf1b2db6236e0e3809562c5078de6b333859bed5a63cbe29f842915ec8f4	1	1	[ "gretel_sql" ]	held_out	curated_candidate	[]	core_candidate	[ "source_sql_complexity:multiple_joins", "sql_structure:selects=1;joins=2;statements=1;advanced=none", "combined_relational_operations_require_calibration" ]	structural_heuristic	Qwen3.6-27B	heuristic_not_model_measured
gretel_sql:eda4970813b44eaa3c5ebb12be71277260d7b86e3cbe8d1927b6601a2fc10028	task:d78605e6064968e795fecfb4b0038b09d61f315911fc55806a42f1e052114882	aeda176cfa04bf7818cf10d1d3f4344108e8dc713ae26e2dce008287cee06ce3	gretel_sql	gretelai/synthetic_text_to_sql	740ab236e64503fba51be1101df7a1be83bf455d	synthetic_text_to_sql_test.snappy.parquet	3651	test	https://huggingface.co/datasets/gretelai/synthetic_text_to_sql/resolve/740ab236e64503fba51be1101df7a1be83bf455d/synthetic_text_to_sql_test.snappy.parquet	apache-2.0	Gretel Synthetic Text-to-SQL dataset creators; source card and author citation preserved in the download manifest.	proposed	What is the number of professional development workshops attended by teachers in each district, grouped by subject area?	CREATE TABLE districts (district_id INT, district_name TEXT); INSERT INTO districts (district_id, district_name) VALUES (1, 'Urban'), (2, 'Suburban'), (3, 'Rural'); CREATE TABLE teachers (teacher_id INT, teacher_name TEXT, district_id INT); INSERT INTO teachers (teacher_id, teacher_name, district_id) VALUES (1, 'Ms. Do...	database_business_operations	database_business_operations.database_query_analysis	heuristic_unreviewed	Mapped from source sql_task_type='analytics and reporting'; no individual SQL execution or semantic review.	en	none	SELECT d.district_name, p.program_subject, COUNT(pd.program_id) as num_programs FROM districts d JOIN teachers t ON d.district_id = t.district_id JOIN professional_development pd ON t.teacher_id = pd.teacher_id JOIN (SELECT program_subject FROM professional_development GROUP BY program_subject) p ON pd.program_subject ...	{"domain":"education","domain_description":"Education data on student mental health, teacher professional development, open pedagogy, and lifelong learning.","id":3651,"sql_complexity":"multiple_joins","sql_complexity_description":"two or more joins (specify inner, outer, cross)","sql_explanation":"We perform a join be...	4195ac8942b8994316b6440bac0c017a7220c83a729c8679da4415735122e87d	gretel_sql:test:3651	multiple_joins	intermediate	Estimated from source sql_complexity='multiple_joins'; query feature labels are only proxies for difficulty, not measured solver performance. No empirical solver calibration was performed.	heuristic	needs_difficulty_calibration	not_measured	analysis_only_unvalidated	primary	task_annotation	3,650	[ "difficulty_not_calibrated", "not_individually_reviewed", "reference_not_independently_verified", "reference_or_constraint_only_no_rubric", "response_diversity_not_validated", "source_evaluation_split", "sql_dialect_and_result_equivalence_need_review", "sql_not_executed", "synthetic_reference" ]	false	true	false	task:d78605e6064968e795fecfb4b0038b09d61f315911fc55806a42f1e052114882	1	1	[ "gretel_sql" ]	held_out	curated_candidate	[]	core_candidate	[ "source_sql_complexity:multiple_joins", "sql_structure:selects=2;joins=3;statements=1;advanced=none", "combined_relational_operations_require_calibration" ]	structural_heuristic	Qwen3.6-27B	heuristic_not_model_measured
gretel_sql:cf8c7ffa0ffb592563557163b5fe7a0e7fc49f6a6801116811e7097e3ed0e526	task:7d95d5bb7f9dd6391a87a09da6169f52bda7f1fa12b4ff9e697f72d5aace4d9e	9666d75526b2e3160d5ca4b49237d2a0ddefc2650d5f2f524a4d638a91d53b75	gretel_sql	gretelai/synthetic_text_to_sql	740ab236e64503fba51be1101df7a1be83bf455d	synthetic_text_to_sql_test.snappy.parquet	3674	test	https://huggingface.co/datasets/gretelai/synthetic_text_to_sql/resolve/740ab236e64503fba51be1101df7a1be83bf455d/synthetic_text_to_sql_test.snappy.parquet	apache-2.0	Gretel Synthetic Text-to-SQL dataset creators; source card and author citation preserved in the download manifest.	proposed	Who are the therapists in Texas that have treated more than 5 patients with anxiety disorder?	CREATE TABLE patients (patient_id INT, patient_name TEXT, condition TEXT, therapist_id INT); CREATE TABLE therapists (therapist_id INT, therapist_name TEXT, state TEXT); INSERT INTO patients (patient_id, patient_name, condition, therapist_id) VALUES (1, 'Oliver Brown', 'Anxiety Disorder', 1); INSERT INTO patients (pati...	database_business_operations	database_business_operations.database_query_analysis	heuristic_unreviewed	Mapped from source sql_task_type='analytics and reporting'; no individual SQL execution or semantic review.	en	none	SELECT therapists.therapist_name FROM therapists JOIN (SELECT therapist_id FROM patients GROUP BY therapist_id HAVING COUNT(*) > 5) AS patient_counts ON therapists.therapist_id = patient_counts.therapist_id WHERE therapists.state = 'Texas' AND therapists.therapist_name IN (SELECT therapists.therapist_name FROM therapis...	{"domain":"mental health","domain_description":"In-depth data on mental health conditions, treatment approaches, patient outcomes, and public awareness campaigns in mental health.","id":3674,"sql_complexity":"multiple_joins","sql_complexity_description":"two or more joins (specify inner, outer, cross)","sql_explanation...	3750c6527d72468e72427dee321fe26631f5ec4b5d9ac6a5e5b2a5cf5d63ef41	gretel_sql:test:3674	multiple_joins	intermediate	Estimated from source sql_complexity='multiple_joins'; query feature labels are only proxies for difficulty, not measured solver performance. No empirical solver calibration was performed.	heuristic	needs_difficulty_calibration	not_measured	analysis_only_unvalidated	primary	task_annotation	3,673	[ "difficulty_not_calibrated", "not_individually_reviewed", "reference_not_independently_verified", "reference_or_constraint_only_no_rubric", "response_diversity_not_validated", "source_evaluation_split", "sql_dialect_and_result_equivalence_need_review", "sql_not_executed", "synthetic_reference" ]	false	true	false	task:7d95d5bb7f9dd6391a87a09da6169f52bda7f1fa12b4ff9e697f72d5aace4d9e	1	1	[ "gretel_sql" ]	held_out	curated_candidate	[]	core_candidate	[ "source_sql_complexity:multiple_joins", "sql_structure:selects=3;joins=2;statements=1;advanced=none", "combined_relational_operations_require_calibration" ]	structural_heuristic	Qwen3.6-27B	heuristic_not_model_measured
gretel_sql:999778bd65ba6ee3c490db9bb1ebad87e213207ae797e75de91d0040f073467f	task:9d5dd19e93d979764548e2ccc7d0971b3109a212bcc7f7bf8e7e13c7c7ea122a	6a39d60de24dc5fa0985425c87b74c14d7d8fcc6984f2d78444e028f76a61e20	gretel_sql	gretelai/synthetic_text_to_sql	740ab236e64503fba51be1101df7a1be83bf455d	synthetic_text_to_sql_test.snappy.parquet	3730	test	https://huggingface.co/datasets/gretelai/synthetic_text_to_sql/resolve/740ab236e64503fba51be1101df7a1be83bf455d/synthetic_text_to_sql_test.snappy.parquet	apache-2.0	Gretel Synthetic Text-to-SQL dataset creators; source card and author citation preserved in the download manifest.	proposed	What is the yield of the top 5 crops in 2021?	CREATE TABLE CropYield (id INT, crop TEXT, year INT, yield REAL);	database_business_operations	database_business_operations.database_query_analysis	heuristic_unreviewed	Mapped from source sql_task_type='analytics and reporting'; no individual SQL execution or semantic review.	en	none	SELECT crop, yield FROM (SELECT crop, yield, ROW_NUMBER() OVER (PARTITION BY crop ORDER BY yield DESC) as rn FROM CropYield WHERE year = 2021) sub WHERE rn <= 5;	{"domain":"agriculture","domain_description":"Comprehensive data on agroecology, food justice, indigenous food systems, and urban agriculture.","id":3730,"sql_complexity":"window functions","sql_complexity_description":"window functions (e.g., ROW_NUMBER, LEAD, LAG, RANk, NTILE, PERCENT_RANK, etc.) with partitioning an...	8d8e79511f1daf450f481d9359867c3a0734fa6997141f2e091b8aa8d63a1edf	gretel_sql:test:3730	window functions	intermediate	Estimated from source sql_complexity='window functions'; query feature labels are only proxies for difficulty, not measured solver performance. No empirical solver calibration was performed.	heuristic	needs_difficulty_calibration	not_measured	analysis_only_unvalidated	primary	task_annotation	3,729	[ "difficulty_not_calibrated", "not_individually_reviewed", "reference_not_independently_verified", "reference_or_constraint_only_no_rubric", "response_diversity_not_validated", "source_evaluation_split", "sql_dialect_and_result_equivalence_need_review", "sql_not_executed", "synthetic_reference" ]	false	true	false	task:9d5dd19e93d979764548e2ccc7d0971b3109a212bcc7f7bf8e7e13c7c7ea122a	1	1	[ "gretel_sql" ]	held_out	curated_candidate	[]	core_candidate	[ "source_sql_complexity:window functions", "sql_structure:selects=2;joins=0;statements=1;advanced=OVER", "combined_relational_operations_require_calibration" ]	structural_heuristic	Qwen3.6-27B	heuristic_not_model_measured

Subsets and Splits

No community queries yet

The top public SQL queries from the community will appear here once available.