- 3kevtiny Jev-like family of decision models built on top of Qwen3.5 you can train and run on your own
jaredpalmer/kev · Python
6,661381jevdecision-modelqwen3python - Native MLX runtime for Laya typed decision models — 7–14 ms short decisions on M3 Max. No text generation, PyTorch, or cloud API.
mizorewww/laya-mlx · Python
6,172473apple-silicondecision-modelinferencelayalocal-aimachine-learningmlxmodernbert - Semantic ifs from open models, on a 3090 at home. Independent; not affiliated with Jev or TypeSafe.
TheoLeeCJ/SemIf-OpenJev · Python
4,206283jevdemoopen-sourcelocal - 15CLMOpen-source contrastive System One decision model with a TypeSafe-compatible API for typed Noul, Choice, and Score judgments.
Contrastive-LM/CLM · Python
63652system-onedecision-modelcontrastive-learningtypesafe-compatibleopen-sourcepython - 21AnyJevIndependent open-source Jev-style decision model that turns open LLMs into typed Choice, Noul, and Score judgments with probabilities.
nokia-applied-research/AnyJev · Python
48072calibrationdecision-modeljevjev-modelllmsystem-onetransformersvllm - 26openjevOpen, Jev-compatible System One decision server on DiffusionGemma
razorback16/openjev · Python
39231jevpython - 31deciderOne-pass typed decisions with calibrated probabilities (System One style model), fine-tuned from Qwen3.5-2B
Mapika/decider · Python
35224jevpython - 35Open-JevOpen-Jev — Jev / TypeSafe related project
Zefan-Cai/Open-Jev · Python
31330jevpython - Open-source alternative to TypeSafe's Jev: a System One style model layer that gives typed, calibrated decisions from any open-weights LLM in one forward pass (HF + vLLM), with honest benchmarks
ikermoel/open-alternative-jev · Python
5311calibrationclassificationjevjev-alternativellmlogprobsopen-jevopen-source-jev - Open-source, local decision models inspired by TypeSafe’s Jev and System One - built on Qwen3 for NVIDIA DGX Spark.
abhishek085/open-spark-jev · Python
153jevagentic-aiapachedatagendgx-sparkharnesslocal-aiopensource - 172snapjudgeTyped decisions (choice / score / yes-no) from local Qwen models on Apple Silicon. Probabilities come straight from the logits, no text generation. TypeSafe-compatible HTTP API, runs on MLX.
Micha0827/snapjudge · Python
130jevapple-siliconclassificationdecision-makingfastapillmllm-evaluationlocal-llm - Jev-style typed decisions from Qwen3.5-2B logits — one forward pass, zero decoding, zero fine-tuning.
Embodied-AI-System/Qwen3.5-OneForward · Python
41fastapijevlogitsqwenqwen3-5self-hostedsystem-onetransformers - 1118typedecideOpen-source implementation of the typed-decision pattern popularised by TypeSafe's Jev: read a decision out of a small language model's logits, in the browser. Library, benchmarks and paper. Not affiliated with TypeSafe.
ardada2468/typedecide · Python
00browser-mlconstrained-decodingfine-tuningjevllm-evaluationllm-inferencelogit-readoutlora