Organization repository on GitHub
Send a state and typed questions, get typed answers with probabilities.
- Forks
- 1
- Last pushed
- Updated Oct 2026
Repo stats from the GitHub API, cached Oct 2026.
Jev directory
CommunityJev Models & Reimplementations
Agr
A decision model from Command Code in two sizes, Agr and Agr-flash: send a state as text or JSON with typed questions and get a probability for every option, with all questions about the same state answered in one forward pass.
- Stars
- 7
- Language
- Python
- License
- APACHE-2.0
- Reading time
- 2 min
- Published by
- Community
- Author
- CommandCodeAI
- Last verified
- Oct 2026
Used inJev for Software & DevTools
- Languages
- Python
- Capabilities
- Open source
Highlights
- Decision model in two sizes, Agr and Agr-flash; the models are collected on Hugging Face under CommandCode/agr.
- Scores 58.15 on Decision Index 0.2.1 across 38 counted benchmarks, per the repository badge.
- Every question about the same state is answered in one forward pass, independently of the others.
- Returns a probability for every option instead of text to parse — routing a ticket, picking a tool, approving an action.
- Ran on all 150,759 requests of the public Decision Index kit; a figure compares it with Jev, Kev, and Rune.
Watch out
README reviewed 2026-10-17; the digest truncates at the end of the evaluation section, so hosting, pricing, weight licences, and usage instructions are not visible. The 58.15 Decision Index score is the project's own badge rather than an independent result.
Questions about this entry
Short answers drawn from the linked project or page.
A decision model: you send it a state, as text or JSON, with one or more typed questions, and it returns typed answers with a probability for every option instead of generating text you have to parse.
Two: Agr and Agr-flash, per the README. The models are hosted in a Hugging Face collection under CommandCode/agr, and the repository badge reports a Decision Index 0.2.1 score of 58.15 across 38 counted benchmarks.
It scores a fixed set of answers directly — routing a ticket, picking a tool, or approving an action — and returns a probability for each, so no generated text has to be parsed. Questions about the same state are answered in one forward pass.
The README says Agr was run on all 150,759 requests of the public Decision Index kit and shows a figure comparing it with Jev, Kev, and Rune on Decision Index 0.2.1. Per-benchmark numbers are not in the digest.
In a Hugging Face collection linked from the repository (huggingface.co/collections/CommandCode/agr). The project badge says Apache 2.0, but the digest truncates before licence details for the weights themselves.
Agr is Command Code's own decision model, benchmarked against Jev and other decision models on the Decision Index. It speaks the same typed-decision pattern — state plus typed questions — rather than being a Jev client or TypeSafe product.
The digest truncates before any usage or serving instructions; the evaluation section ends with the Decision Index kit link. The Hugging Face collection is the natural starting point for weights and inference details.
More Jev resources like this
An open decision model built on frozen Qwen3.8-27B that processes a shared application state with isolated question branches and returns probabilities with zero generated answer tokens. It reports 87.45% on JevBench, exceeding Open-Jev-27B-v1.1 by 2.16 points, and needs 64 GiB of free GPU memory.
Jev Models & Reimplementations#community#github#python
Community
Camellia86/Canopy-JevPython
Camellia86/Canopy-JevPythonThree open-weights 4B decision models for English (Deem-4B), Turkish (Karar-4B) and German (Wähler-4B) that choose between options and return a probability for each. The server implements the Jev API, POST /v1/systemone, so existing Jev clients work by changing the endpoint.
Jev Models & Reimplementations#community#github#python
Community
mertkayacs/jevaltPython
mertkayacs/jevaltPythonA self-trainable 2B browser-agent decision model that speaks TypeSafe's /v1/systemone protocol, so an existing jev-ultrafast agent is repointed with a one-line base-URL change. Runs as a 1.6 GB Q4_K_M GGUF under llama.cpp at about 236 ms per decision.
Jev Models & Reimplementations#community#github#python
Community
chemany/jevaPython
chemany/jevaPythonA small, fast decision model — a system one model — for agentic workflows from Strands Labs: it picks between options or rates on a scale faster than an LLM and returns a calibrated confidence with every decision.
Jev Models & Reimplementations#community#github#python
Community
strands-labs/strands-deciderPython
strands-labs/strands-deciderPythonAn open-source, local alternative to Jev: System-One decision models that answer the POST /v1/systemone shape — choice, yes/no, or score — for general decisions and browser-agent steps. Models ship as OpenJev-1.7B, 4B, and 8B, with a pip-installable Python package.
Jev Models & Reimplementations#community#github#python
Community
alanhuangyoo/openjevPython
alanhuangyoo/openjevPythonA Chinese-language decision model that returns a probability for each fixed candidate option in one forward pass, with no generated text or chain of thought. It is a LoRA fine-tune of Qwen3.5-0.8B on 13,048,000 CJ-Dataset rows, benchmarked on CJ-Bench against Chinese-Jev.
Jev Models & Reimplementations#community#github#python
Community
yanqiangmiffy/chinese-decisionPython
yanqiangmiffy/chinese-decisionPythonFrom 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.