kyotsu-ai-bench
Static benchmark dashboard comparing Jev with two OpenAI Luna settings on Japan's 2026 university Common Test.
This summary is based on how curated lists describe the project; its README was not available to review.
Overview
kyotsu-ai-bench is an AI benchmark built on Japan's 2026 Common Test for University Admissions, presented as a static dashboard. It compares Jev against 'luna-none' and 'luna-low', which appear to be an OpenAI Luna model at two reasoning-effort settings, on the exam's multiple-choice questions.
No README was available, so the subjects covered, scoring method and results are not confirmed here. It is relevant as one of the few benchmarks of Jev on Japanese-language exam questions.
What the decision model decides
| Type | Question | Answers |
|---|---|---|
| Choice | Which answer option is correct for this Common Test question?(inferred) | — |
How it works
- Convert exam questions to multiple-choice format
- Run Jev and the comparison models
- Score answers against the official key
- Publish results in a static dashboard
Strengths
- Japanese-language benchmark for Jev
Caveats
- No README available to verify method or results
- Model naming for comparisons is unclear
At a glance
- Type
- Benchmark
- Status
- research
- Runtime
- static dashboard
Use cases and integrations
How curated lists describe it
AI benchmark on Japan's 2026 Common Test: Jev vs luna-none vs luna-low (static dashboard).
hellogumbo/awesome-jev
Frequently asked questions
What does it benchmark?
Jev against two Luna settings on Japan's 2026 Common Test.
Are results available?
They are presented on a static dashboard; details were not available here.
Which language are the questions in?
Japanese.
Related projects
- jev-arena★ 122
Side-by-side tool comparing Jev and DeepSeek on labelling the same comments for relevance, sentiment and intent, with replay and reports.
ProjectbetaTypeScript Evaluation suite for decision models with 23,900 pinned rows across nine task families and three primitives, plus a leaderboard.
ProjectbetaPythonOpen-source BYOK arena where Jev faces another model as a judge on your question, with blind voting before costs are revealed.
ProjectbetaTypeScriptpytest-style tool that pins fixtures and expected Jev answers, then fails a model upgrade when answers flip or confidence drops.
ProjectbetaPythonTest harness for any Jev-API decision model: accuracy, calibration and robustness probes for reordering, injection and padding.
ServicebetaPythonSlack-like playground for watching computer-use agents work, with replays, comparable runs and Jev choosing from a menu of visible actions.
ProjectbetaTypeScript