Open decision models and reproductions with decision models

Zero-dependency npm package that runs Laya natively or as WebAssembly in Node, Bun and browsers, drop-in for Jev's /v1/systemone.

Native C++ Laya inference on ggml with CUDA, Vulkan and Core ML, plus a Jev-compatible HTTP server with automatic batching.

Turns any chat model on SGLang, vLLM, transformers or MLX into a Jev-compatible service with one prefill and one label token per question.

Docker recipes and ARM64 images for serving ten native decision checkpoints, such as Cloudflare Clef, behind /v1/systemone on DGX Spark.

NoulXP: open standard and reference engine for packaging calibrated System One decision models so one runtime can run them all.

Local-first desktop AI workbench for llama.cpp, vLLM and SGLang that includes a JEV typed-decision page and /v1/systemone gateway.

Japanese System One decision model on ModernBERT-ja (314.6M) returning typed choice, score and bool distributions with no generation.

Self-hosted Jev-compatible System One server that sits in front of llama-server and reads option probabilities from any GGUF model.

Korean RL tycoon game training a company decision model that answers situations with choice, probabilities and confidence.

Valence: macro news sentiment engine scoring global headlines' market impact with a contrastive System-One model on serverless GPU.

MLX library that turns an existing local model into a decision engine by reading candidate logits, with no fine-tuning, on Apple Silicon.

Experiment running a trained jevlike option scorer as ESP-IDF firmware on an ESP32 microcontroller.

Early Rust runtime that answers Jev and Laya typed questions, currently serving Laya checkpoints on CPU, CUDA and Apple GPUs.

Browser-only Jev-style decisions in Chinese or English using a zero-shot NLI model on ONNX Runtime Web, with honest limits.

Open System One-style playground: send state plus typed choice, score and noul questions to any of four backends via a UI or API.

Jev-compatible open System One engine with a Claude Code-style terminal REPL showing live probabilities, plus an experiment harness.

Python library that turns an open LLM into a Jev-style classifier via constrained single-token choice indexes in one forward pass.

Browser arcade games (2048, runner, chess, Flappy Bird, Sudoku) where small decision models choose each move from text options.

Open Jev-style decision model built on Gemma 3 270M with a scoring head, scoring all options in one forward pass on CPU.

Self-hosted encoder model giving Jev-compatible choice, score and noul answers in one pass, pitched as a better-calibrated Laya successor.

33-chapter Chinese tutorial on agentic RL and GRPO, including a chapter on when Jev-style external decision models beat training with RL.

Joke decision model trained to answer wrongly with total confidence: 1.9% accuracy at 0.957 mean confidence, via a Jev-style API.

Open clinical text classifiers (0.8B-9B on Qwen3.5) with labels set at inference time, run locally and compared with Jev 1.13.

Pre-registered field test of Jev, GLiNER2.5-Decide, Laya and other decision models against small LLMs on fresh September 2026 data.

Open 4B System I driving model that scores behaviours like cruise, stop, yield and brake from three camera frames each decision.

11 MB CPU-only model that answers choice, noul and score questions with fitted calibration and no LLM at inference.

Official code for the FastJEV paper: training-free reduction of redundant candidate computation in multimodal JEV-style models.

Tiny pre-registered experiment testing whether a confidence-scoring reward (RLCD-like) yields confidences safe to automate on at 0.9.

Research code for an image-and-text System One model that reads option-letter logits from a multimodal LLM, with training and calibration.

Instruction dataset of 119 rows compiled from TypeSafe's public Jev documentation, split by Choice, Noul and Score.