Upload folder using huggingface_hub

README.md

@@ -12,11 +12,12 @@ license: mit
 
 # Helium Beat-48 Challenge
 
-Can your LLM read a real option chain? Sample frozen prompts from the [Market Resolution benchmark](https://huggingface.co/datasets/HeliumTrades/helium-market-resolution-benchmark) **v3.0** and compare to the current frontier leaderboard.
+Can your LLM read a real option chain?
 
-v3 scoring covers new tasks (`time_value`, `parity_prices`, `term_structure_table`, `chain_surface_mcq`) and drops answer-leak term-structure MCQs.
+Sample frozen prompts from the [Market Resolution benchmark](https://huggingface.co/datasets/HeliumTrades/helium-market-resolution-benchmark) (v3) and compare to the current frontier.
 
-Set `OPENAI_API_KEY` (or other provider keys) in Space secrets. Uses LiteLLM model strings.
+Set provider keys in Space secrets. Uses LiteLLM model strings.
 
-- [Landing page](https://heliumtrades.com/benchmarks/)
-- [Model Worldview benchmark](https://huggingface.co/datasets/HeliumTrades/helium-model-worldview-benchmark)
+- [Market Resolution](https://huggingface.co/datasets/HeliumTrades/helium-market-resolution-benchmark)
+- [Model Worldview](https://huggingface.co/datasets/HeliumTrades/helium-model-worldview-benchmark)
+- [Landing](https://heliumtrades.com/benchmarks/)