undrgrnd-labs/hunch

Local proxy that learns your app's typed LLM decisions and answers them with a Laya head. Jev and OpenAI compatible.

Open infrastructure for training, evaluating, and deploying System 1 decision models across language and multimodal backbones.

Open, calibrated System 1 decision models: typed choice / score / yes-no answers with a probability for every option. 400M to 80B; CPU, NVIDIA, Apple Silicon, LM Studio, Ollama and vLLM.

Turn any LLM into a Jev-style decision model: typed decisions, real probabilities, no training. (continue updating, welcome any issue and PR request)

Jev-like family of decision models built on top of Qwen3.5/3.8 you can train and run on your own

A family of System One-style models fine-tuned from Qwen3.5, designed for one-pass typed decisions with calibrated probabilities.

Millisecond decisions, any domain: a 0.8B open "System 1" model that picks between your options with calibrated probabilities. One base, swappable LoRA adapters, on your own hardware.

Distill your LLM traces into a fast hybrid System 1 models. Open-source ML + LLM inference with calibrated deferral.