Instructions to use codelion/sprog-9m with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- MLX
How to use codelion/sprog-9m with MLX:
# Download the model from the Hub pip install huggingface_hub[hf_xet] hf download codelion/sprog-9m --local-dir sprog-9m
- Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- LM Studio
- Atomic Chat
Commit History
Yesterday
-
Remove pipeline_tag (question-answering means extractive QA on the Hub; this is a program-generating solver)
Verified signaturee44d22b
-
Clean up model card prose (remove em-dashes and inline bold emphasis)
Verified signature625f50d -
Fix model-index dataset type: gsm8k -> openai/gsm8k (canonical Hub id)
Verified signatureb36266d
-
Add GSM8K eval result on main with date (leaderboard linking)
Verified signature0311161 -
Move GSM8K eval result to community PR
Verified signature0ca1ac3 -
Add model-index GSM8K result for right-side eval widget
Verified signature8d6dc47 -
Add GSM8K eval result for leaderboard integration (self-reported, symbolic verifier)
Verified signature6d52fea
-
Clarify metric: verifier@96 self-consistency (single answer), not pass@96
Verified signatureec3988b -
Fix strikethrough rendering (tildes -> approx symbol)
Verified signaturebf2f12f -
Fix pipeline_tag metadata
Verified signature758ab1c -
Add SPROG-9M: 9.37M from-scratch LLM-free GSM8K solver (~11.8% test)
Verified signature1a4b406 -
initial commit
Verified signature4874d05