Jev directory
CommunityJev Models & Reimplementations
OpenJev on Docker Hub
Docker's verified ai/openjev image: an 18.9 GB artifact converting the OpenJev weights from huggingface.co/openjev/openjev, served with llama serve through the /v1/systemone API and released under CC BY-NC 4.0 for non-commercial use only.
- Reading time
- 2 min
- Published by
- Community
- Author
- Docker (ai/openjev)
- Last verified
- Oct 2026
- Platforms
- API
- Capabilities
- Open sourceSelf-hostedRuns locally
Highlights
- Verified-publisher Docker image ai/openjev whose 18.9 GB artifact converts the OpenJev weights from huggingface.co/openjev/openjev.
- Runs with llama serve -hf ggml-org/OpenJev-GGUF, using the /v1/systemone decision API instead of a text-generation endpoint.
- Weights are released under CC BY-NC 4.0, non-commercial use only, per the repository note.
- Converted automatically with the ggml-org/convert tooling, and the page points to ggml-org/llama.cpp pull request 29818 for the decision-model support.
- Listing shows the latest tag updated four days before capture, 944 pulls that week, and a digest beginning sha256:e22193ece.
Watch out
The licence is non-commercial: CC BY-NC 4.0 forbids commercial use of the weights per the repository note, and only the upstream Hugging Face repository is named as the source. 18.9 GB is a large pull, no accuracy benchmarks or context length appear on the page, and automatic conversion means the image follows upstream rather than publishing independent evaluations.
Questions about this entry
Short answers drawn from the linked project or page.
A verified-publisher Docker image under the ai namespace that packages OpenJev, a decision model, for use through the /v1/systemone API. The 18.9 GB artifact was converted from the weights at huggingface.co/openjev/openjev.
The page shows llama serve -hf ggml-org/OpenJev-GGUF, plus docker pull ai/openjev for the container itself. The model is a decision model to be called through /v1/systemone rather than a chat endpoint.
CC BY-NC 4.0, which the repository note flags as non-commercial use only. That restricts commercial deployment regardless of where the container runs, and the page provides no alternative licensing route.
Automatically, using the ggml-org/convert tooling, with the page pointing to ggml-org/llama.cpp pull request 29818 for decision-model support. Because it is automated, upstream weight changes flow through without hand-editing.
18.9 GB for the latest tag, with a digest beginning sha256:e22193ece. The page shows 944 pulls in the week before capture, and the tag was last updated four days before capture.
No. The page is a Docker Hub listing with no readme benchmarks, no context length, and no accuracy claims; the only description is that it is a decision model used via the /v1/systemone API.
The ai namespace as a Verified Publisher, listed as by Docker. The underlying weights come from the openjev organisation on Hugging Face, and the GGUF build invoked by llama serve is hosted at ggml-org/OpenJev-GGUF.
Reactions & coverage
Posts, threads, and videos about this entry from around the web.
X: Standard One SH (StandardThinking)
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 NewsMore Jev resources like this
TurinTech's independent GGUF conversion of the 27B Kev decision model for llama.cpp: the default IQ4_XS file is 13.91 GB and keeps 97.8% of the BF16 original's answers over 2,000 test decisions, with zero changed confident answers.
Jev Models & Reimplementations#community#open-models#local
Community
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++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/verdictPythonA LoRA plus 24-slot decision head over Qwen3.8-27B from Ming Xu's train-jev project: it serves TypeSafe's System One API with 0.851 Kev transfer-v4 accuracy and ECE 0.053, and needs an 80 GB GPU.
Jev Models & Reimplementations#community#open-models#local
Community
A 24 GB-GPU alternative from the train-jev project: Qwen3.5-4B plus a rank-16 LoRA and 24-slot decision head scoring 0.647 PK accuracy, 40.4 Decision Index and 43 ms p50 latency, ahead of Kev-4B at the same size.
Jev Models & Reimplementations#community#open-models#local
Community
A CPU-first decision model from Decision Models that answers the seven questions an LLM auto-router needs (category, difficulty, stakes, four flags) in one ONNX pass; MIT weights plus a Jev-compatible hosted API.
Jev Models & Reimplementations#community#open-models#local
Community
From the Jev guides
Original write-ups on what this entry is for.
From the Jev AI model community
Posts from builders shipping with Jev right now.