Yinsongxu/LLM2Jev
VerifiedLocal alternativeDocumented
LLM2Jev (Yinsongxu) turns local text and vision models into Jev-style decision services through prefill scoring, with SGLang, Transformers and MLX backends.
LLM2Jev scores candidates from local language and vision models using prefill logits without token-by-token generation.
It provides an independent Jev-style implementation with a compatible System One HTTP endpoint.
Use Python or HTTP with SGLang, Transformers or MLX, including supported text-and-image inputs.
Cache benefits depend on input length and candidate count; the cited JevBench comparison covers 231 public items.
Evidence checked:
Turn local language models into Jev-style structured decision models. Get results from text and images with prefill alone—no token-by-token decoding required.
Stars measure the whole repository, including work unrelated to Jev.
Documented records the linked documentation or source. JevHunt has not independently run or benchmarked this project.
Evidence commit: a4aafa85e1aba95329db36e085373d1572ab6917
Discovered through: awesome-jev, typesafe-field-guide, github-search.
How we review →Suggest a correction · Claim this listing · Submit a project