Instructions to use jkim96/Qwen3.6-27B-DASHQ-INT3-g128 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use jkim96/Qwen3.6-27B-DASHQ-INT3-g128 with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("image-text-to-text", model="jkim96/Qwen3.6-27B-DASHQ-INT3-g128", trust_remote_code=True)
messages = [
    {
        "role": "user",
        "content": [
            {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"},
            {"type": "text", "text": "What animal is on the candy?"}
        ]
    },
]
pipe(text=messages)

# Load model directly
from transformers import AutoProcessor, AutoModelForCausalLM

processor = AutoProcessor.from_pretrained("jkim96/Qwen3.6-27B-DASHQ-INT3-g128", trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained("jkim96/Qwen3.6-27B-DASHQ-INT3-g128", trust_remote_code=True, device_map="auto")
messages = [
    {
        "role": "user",
        "content": [
            {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"},
            {"type": "text", "text": "What animal is on the candy?"}
        ]
    },
]
inputs = processor.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(processor.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use jkim96/Qwen3.6-27B-DASHQ-INT3-g128 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "jkim96/Qwen3.6-27B-DASHQ-INT3-g128"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "jkim96/Qwen3.6-27B-DASHQ-INT3-g128",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

Use Docker

docker model run hf.co/jkim96/Qwen3.6-27B-DASHQ-INT3-g128

SGLang

How to use jkim96/Qwen3.6-27B-DASHQ-INT3-g128 with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "jkim96/Qwen3.6-27B-DASHQ-INT3-g128" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "jkim96/Qwen3.6-27B-DASHQ-INT3-g128",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "jkim96/Qwen3.6-27B-DASHQ-INT3-g128" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "jkim96/Qwen3.6-27B-DASHQ-INT3-g128",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

Docker Model Runner
How to use jkim96/Qwen3.6-27B-DASHQ-INT3-g128 with Docker Model Runner:
```
docker model run hf.co/jkim96/Qwen3.6-27B-DASHQ-INT3-g128
```

jkim96 commited on Jun 17

Commit

20348e0

verified ·

1 Parent(s): 1b7b00c

Standardize model card (DASH-Q)

Browse files

Files changed (1) hide show

README.md +15 -32

README.md CHANGED Viewed

@@ -1,19 +1,23 @@
 ---
 license: apache-2.0
 base_model: Qwen/Qwen3.6-27B
 library_name: transformers
 tags:
 - dashq
 - quantized
 - post-training-quantization
-base_model_relation: quantized
 ---
 # Qwen3.6-27B-DASHQ-INT3-g128
-This repository contains a DASH-Q packed quantized checkpoint for `Qwen/Qwen3.6-27B`.
-DASH-Q checkpoints require the lightweight DASH-Q runtime package for loading. They are not plain Transformers checkpoints because linear layers are stored as `PackedQuantizedLinear` modules.
 ## Install
@@ -26,10 +30,7 @@ pip install git+https://github.com/JaeminK/dashq.git
 ```python
 from dashq import load_quantized
-model, tokenizer = load_quantized(
-    "jkim96/Qwen3.6-27B-DASHQ-INT3-g128",
-    device_map="auto",
-)
 ```
 ## Quantization
@@ -37,30 +38,12 @@ model, tokenizer = load_quantized(
 | Field | Value |
 | --- | --- |
 | Base model | `Qwen/Qwen3.6-27B` |
-| Bits | `3` |
-| Group size | `128` |
-| Scale/zero dtype | `float16` |
-| Calibration dataset | `wikitext2` |
-| Calibration samples | `128` |
-| Sequence length | `2048` |
-| Original size | `55.5630 GB` |
-| Quantized size | `16.5138 GB` |
-## Evaluation
-| Metric | Value |
-| --- | ---: |
-| `wikitext2_ppl` | 7.7554 |
-| `zero-shot accuracy avg` | 70.8034 |
-| `arc_challenge` | 59.1297 |
-| `arc_easy` | 75.4630 |
-| `commonsense_qa` | 86.0770 |
-| `gsm8k_cot` | 93.1766 |
-| `hellaswag` | 82.6429 |
-| `lambada_openai` | 74.9078 |
-| `mmlu` | 85.8923 |
-| `openbookqa` | 45.2000 |
-| `piqa` | 82.2633 |
-| `truthfulqa_mc2` | 54.8307 |
-| `winogrande` | 76.7167 |

 ---
 license: apache-2.0
 base_model: Qwen/Qwen3.6-27B
+base_model_relation: quantized
 library_name: transformers
 tags:
 - dashq
 - quantized
 - post-training-quantization
+- int3
 ---
+![DASH-Q](https://raw.githubusercontent.com/JaeminK/dashq/main/assets/dashq_banner.png)
 # Qwen3.6-27B-DASHQ-INT3-g128
+> **DASH-Q** — Diagonal-Aware Shrinkage for Robust PTQ.
+> `INT3` · group size 128 · **16.5138 GB** (from 55.5630 GB — **3.4x smaller**)
+DASH-Q checkpoints load with the lightweight DASH-Q runtime — linear layers are packed `PackedQuantizedLinear` modules, not plain Transformers weights.
 ## Install
 ```python
 from dashq import load_quantized
+model, tokenizer = load_quantized("jkim96/Qwen3.6-27B-DASHQ-INT3-g128", device_map="auto")
 ```
 ## Quantization
 | Field | Value |
 | --- | --- |
 | Base model | `Qwen/Qwen3.6-27B` |
+| Precision | INT3, group size 128 |
+| Scale / zero dtype | float16 |
+| Calibration | wikitext2, 128 samples x 2048 |
+| Size | 16.5138 GB  ·  original 55.5630 GB  ·  3.4x compression |
+## Benchmarks
+Full zero-shot / few-shot results for every DASH-Q checkpoint:
+**[github.com/JaeminK/dashq#benchmarks](https://github.com/JaeminK/dashq#benchmarks)**