User repository on GitHub
Semantic ifs from open models, on a 3090 at home. Independent; not affiliated with Jev or TypeSafe.
- Forks
- 348
- Last pushed
- Updated Sep 2026
Repo stats from the GitHub API, cached Oct 2026.Homepage
CommunityJev Models & Reimplementations
SemIf-OpenJev
An open-model semantic-if engine positioned as a local Jev-style decision stack, separate from the original SemIf repo but sharing the same author.
- Stars
- 4.8k
- Language
- Python
- License
- MIT
- Published by
- Community
- Author
- TheoLeeCJ
- Last verified
- Oct 2026
About this project
Semantic ifs from open models, on a 3090 at home. Independent project; not affiliated with Jev or TypeSafe. Wow! No waitlist. Run it in your browser today. Same frozen 4B model · same state · same 21 questions · measured separately, aligned at t=0 in the replay. Most agent decisions are small: route this, retry that, does the evidence support X? A chat model can answer them, but it spends time generating text that software immediately parses back into an if statement.
Quoted from the project's README. Read the full README.
- Languages
- Python
- Capabilities
- Open sourceRuns locally
Highlights
- The README tagline targets semantic ifs from open models running on a home GPU.
- MIT and explicitly not affiliated with TypeSafe or Jev per the repository description.
- Complements the existing SemIf entry as a sibling OpenJev-oriented fork line.
Watch out
Independent weights and API; do not assume parity with hosted Jev calibration or latency claims.
Questions about this entry
Short answers drawn from the linked project or page.
The README served for this repository is the renamed SemIf one, including the line saying the project was formerly called OpenJev and is unaffiliated with TypeSafe. Nothing in it describes a distinct engine, so treat the two listings as one line of work until the repos actually diverge.
It reproduces Jev's interface pattern with open models, not Jev's model or training. On the 102 rows that could be aligned from public artifacts, direct-logit readout agrees with Jev's published 0.883 at 0.845; the README notes no live Jev endpoint was ever called.
Python 3.10 plus one of three backends: a CUDA GPU that can hold a 4B BF16 model for the Torch path, a local GGUF checkpoint for the CPU llama.cpp backend, or an MLX or PyTorch MPS backend on Apple Silicon. Weights are downloaded, not bundled.
Not always. The README labels the reuse paths experimental: BF16 execution changed 5 to 6 of 777 argmaxes against fresh scoring. Parallel suffixes reach 20.03 decisions per second versus 2.33 for fresh direct scoring, so you trade a little determinism for the speed.
Use Python 3.10+ with CUDA and a GPU that holds a 4B BF16 model: create a venv, pip install -e '.[test]', then run semif-score with --mode direct and your input JSONL. Apple Silicon can use the MLX backend or PyTorch MPS, and a CPU-only machine can use the llama.cpp backend with a GGUF checkpoint.
On the same frozen Qwen3.5-4B, state and 21 binary criteria on one RTX 3090, reading typed logits took 1.023 s with zero output tokens, while generating a compact ordered yes and no JSON array took 5.332 s and 111 output tokens. Completing the array was 5.21x as long, and the choices agreed with direct argmax on 18 of 21 criteria.
On the 102 rows that could be aligned from public artifacts, Qwen3.5-4B direct logits reached 0.845 modal agreement with TypeSafe's selected subset, against published Jev's 0.883. The Jev number is read from TypeSafe's published records rather than a live endpoint, and the comparison does not cover TypeSafe's reported 711-row aggregate.
Reactions & coverage
Posts, threads, and videos about this entry from around the web.
LinkedIn: What multimodal decision models actually change, and what they do not.
LinkedInJevDirectory.org
Hacker News: Decision Index 0.2.1 scores open reproductions of Jev on a frozen suite.
jackbravo4 pointsDecision Index 0.2.1 – 70 open reproductions of TypeSafe Jev's Decision ModelRead the thread on Hacker NewsHacker News: Pointer to a DiffusionGemma plus vLLM tutorial for a Jev-shaped classifier.
1 pointHow to build a Jev-style classifier with DiffusionGemma and vLLMRead the thread on Hacker NewsHacker News: A one-token logprob wrapper that also classifies webcam frames.
allanrbo151 pointsA single function Jev-like wrapper for LLMs, including vision modelsRead the thread on Hacker NewsMore Jev resources like this
1.3kGitHub stars
A local yes/no decision model that speaks Jev's wire format on a laptop CPU, refusing Choice and Score until those question types ship.
Jev Models & Reimplementations#community#python#open-models
Community
feder-cr/jevC++
feder-cr/jevC++8GitHub stars
A Python server in front of llama-server that reads one-token label probabilities and exposes them as POST /v1/systemone.
Jev Models & Reimplementations#community#python#open-models
Community
khimaros/verdictPython
khimaros/verdictPython6.9kGitHub stars
A native MLX runtime for Laya's open typed-decision weights: short decisions on Apple Silicon, with no generated tokens, PyTorch, or cloud API.
Jev Models & Reimplementations#community#python#open-models
Community
mizorewww/laya-mlxPython
mizorewww/laya-mlxPython1.6kGitHub stars
Laya typed decisions through Apple Core ML and the Neural Engine, with the author's speed and energy benchmarks for short decisions on an M3 Max.
Jev Models & Reimplementations#community#python#open-models
Communitymizorewww/laya-coremlPython
mizorewww/laya-coremlPython877GitHub stars
A local server that reads next-token probabilities from a fine-tuned Spark model and returns Choice, Score, and Noul answers with zero generated tokens.
Jev Models & Reimplementations#community#python#open-models
Community
Rizzo-AI-Academy/rizzo-flowPython
Rizzo-AI-Academy/rizzo-flowPython738GitHub stars
A multimodal decision model that scores text, image, and video candidates with a shared head on Qwen3.5, plus SFT and experimental RLCD training code.
Jev Models & Reimplementations#community#python#open-models
Community
Liuziyu77/ValenPython
Liuziyu77/ValenPythonFrom the Jev guides
Original write-ups on what this entry is for.
Taking a decision checkpoint out of Python: what the GGUF and browser WebGPU exports preserve, what they change, and the parity evidence the model cards publish.
Jev Tools & CLIs#open-models#local#onnx
Read article
Interface, size, license, claims — in that order. A selection guide across the system-one catalog, from a 706K form scorer to 27B Jev students.
Jev Practices & Patterns#open-models#evaluation#benchmarks
Read article
Hundreds of repositories carry the system-one tag, but the ecosystem is a dozen ideas in different sizes. A tour of the families, the licenses, and what a download does not include.
Jev Models & Reimplementations#open-models#hugging-face#local
Read article
From the Jev community
Posts from builders shipping with Jev right now.