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.
Yinsongxu/LLM2Jev: 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.
