GPT-AGI/OpenJev

The open, local take on Jev: typed decisions from an LLM, without generating a single token

Super fast Jev like server in JS, model agnostic, working with any OpenAI compatible endpoint

Open decision models from 0.8B to 27B: a calibrated probability for every option in one pass, and a short thought only when unsure. Engine for vLLM, llama.cpp and MLX.

OpenSource Jev

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.

Jev-compatible API endpoint based on open models (prefill-only)

Turn any off-the-shelf LLM into a Jev -like decision layer

Open Drex decision models: runtime, docs and model pages

Local decision model with calibrated probabilities: send a state and yes/no, choice or score questions, get a probability for every option. 0.8B GGUF on CPU, Jev-style API.