Benchmarks / Kai / Release gate against Jev
Kai · Release gateRelease gate against Jev
Whether a Kai revision regresses past tolerance against Jev on any gated suite. Passing it is the condition for calling a revision state of the art.
Latest
History
4 results, newest first. Superseded runs stay, so a revision's progress can be read.
| Measured | Subject | Metric | Value | Baseline | Result | Status |
|---|---|---|---|---|---|---|
| Sep 28, 2026 | Kai a8 · 0cfeef05 | Gate result | fails · 14 regressed of 62 suites | – | Loss | Live · Hanzo-measured |
| Sep 28, 2026 | Kai a7 · 0834a74f | Gate result | fails · 13 regressed of 62 suites | – | Loss | Superseded · Hanzo-measured |
| Sep 28, 2026 | Kai a6 · a211cc70 | Gate result | fails · 11 regressed of 62 suites | – | Loss | Superseded · Hanzo-measured |
| Sep 27, 2026 | Kai a5 · df1c16a2 | Gate result | fails · 16 regressed of 62 suites | – | Loss | Superseded · Hanzo-measured |
Conditions
tolerance accuracy 0.02, ece 0.03Hardware
The run records no device or host.
Source
research run gate.jev-drawn-a8.0cfeef05-kai
Reproduce
No public command regenerates this result. The source above is the record it is read from.
More Kai benchmarks
AG News · DAIR Emotion · Banking77 · Support triage · Email spam · Phishing · Jailbreak · Toxicity · RAG relevance · Model routing · Typed decisions · MASSIVE · AG News, validation split · Support triage, validation split · Typed decisions, validation split · HarmBench (benign prompts) · Choosing among many options · Joint decisions · Latency of one decision
Every result for Release gate against Jev on the shelf · How these are measured