NanmiCoder/jev-arena
Jev 模型介绍与实测：通过 Choice / Score / Noul 将自然语言转为带类型的判断与概率，用于分类、评分和路由；支持与 DeepSeek 等模型对比评论打标、速度与结果，含 CSV/Excel 导入、原速回放与离线报告。
About the project
A web tool for comparing two LLMs on the same labeling task: speed, cost and accuracy on CSV/Excel comment data, with run recording and offline reports.
Useful for
- Compare Jev and DeepSeek speed and cost on your own CSV of comments
- Check two models' labeling accuracy on relevance, sentiment and intent
- Generate an offline HTML report from a finished run without API calls
README summarized by DeepSeek V4.1 Flash. Details may be inaccurate.
Why it’s trending
- 0 stars so far today, about 2 expected by the end of the day.
- The repository is 5 days old and already has 97 stars. With less than two weeks of history, there's no usual pace to compare the spike against yet.
- Top new repositories this week: #172.
Stars per day
Bars are daily stars, the line is the usual pace. Red marks spike days.
Numbers
- Total stars
- 97
- Today
- 0 · ≈ 2 by evening
- Forks
- 9
- Issues and pull requests
- 1
- Watchers
- 0
- Language
- JavaScript
- License
- MIT
- Created
- September 19, 2026
- Last push
- September 20, 2026
Star trust
Growth looks organic: forks and discussion are in line with active projects, and stars arrive unevenly, the way people give them.
These are heuristics, not a verdict: we judge by the repository’s behavior, not by a list of stargazers.
Spotted in
- September 24, 2026Top new repositories this week: #170
Share
README badge
Paste it into your README — the badge shows the current magnitude and links to this page.
[](https://gitnova.dev/en/r/NanmiCoder/jev-arena)Similar by description
-
2.3
zhulinchng/jevper
A classification wrapper over OpenAI-compatible clients: instead of prose it returns typed questions (noul, choice, score) with probabilities and confidence. Works with any client exposing responses.create or…
-
2.4
fstandhartinger/jevbench
A benchmark for Jev-class decision models: the model receives state and a rubric, returns a typed answer with probabilities. It scores accuracy, calibration, speed, and cost.