Decisions, not completions.
Kai returns a calibrated choice, score or abstention for the question you state.
Example: a scripted scenario, not model output. Switch a source off and only the decisions that read it run again.
Typed answers
Answers software can act on.
Each one typed, with its probability. When the evidence falls short, Kai says what it needs.
Inside Hanzo
It steers Hanzo’s own agents.
Hanzo Dev, Hanzo Bot and Hanzo Team call Kai for the model, the tools, the context and when to stop.
Results
Kai leads Jev on mean accuracy across 12 suites.
85.7% mean accuracy across 12 suites, against Jev's 78.0%. Ahead of Jev on 11 of 12 suites, behind on 1.
85.7%
Mean accuracy, 12 suites
Jev 78.0%
11 of 12
Suites won vs Jev
AG News, DAIR Emotion, Banking77, Support triage, Email spam, Phishing, Toxicity, RAG relevance, Model routing, Typed decisions, MASSIVE, 51 languages
11 · 0 · 1
Won · tied · behind vs Jev
behind by points on Jailbreak -2.2
8 of 12
Suites where Kai is better calibrated
lower expected calibration error than Jev
30 of 51
MASSIVE languages above Jev
macro accuracy 91.0% against 89.0%
Jev is typesafe/jev-1.13-20260917 through OpenRouter.
| Accuracy | Calibration error (ECE) | |||
|---|---|---|---|---|
| Suite | Kai | Jev | Kai | Jev |
| AG News | 94.3% | 86.0% | 0.033 | 0.102 |
| DAIR Emotion | 93.5% | 60.3% | 0.018 | 0.280 |
| Banking77 | 90.8% | 83.5% | 0.074 | 0.073 |
| Support triage | 44.0% | 36.5% | 0.444 | 0.483 |
| Email spam | 99.8% | 97.8% | 0.003 | 0.060 |
| Phishing | 99.0% | 90.0% | 0.010 | 0.039 |
| Jailbreak | 91.8% | 94.0% | 0.074 | 0.047 |
| Toxicity | 80.3% | 66.3% | 0.195 | 0.177 |
| RAG relevance | 68.0% | 62.0% | 0.028 | 0.279 |
| Model routing | 100.0% | 97.7% | 0.000 | 0.014 |
| Typed decisions | 75.9% | 73.6% | 0.171 | 0.047 |
| MASSIVE, 51 languages | 91.0% | 89.0% | 0.060 | 0.064 |
| Mean, 12 suites | 85.7% | 78.0% | 0.093 | 0.139 |
The same questions to both. MASSIVE is the macro over 51 languages. Kai measured on CUDA BF16; production answers on CPU.
Every language
| Accuracy | ||
|---|---|---|
| Language | Kai | Jev |
| Afrikaans | 90.0% | 87.0% |
| Amharic | 85.0% | 85.0% |
| Arabic | 92.0% | 87.0% |
| Azerbaijani | 93.0% | 88.0% |
| Bangla | 92.0% | 86.0% |
| Welsh | 85.0% | 76.0% |
| Danish | 92.0% | 88.0% |
| German | 91.0% | 93.9% |
| Greek | 94.0% | 94.0% |
| English | 94.0% | 92.0% |
| Spanish | 92.0% | 93.0% |
| Persian | 94.0% | 95.0% |
| Finnish | 93.0% | 89.0% |
| French | 91.0% | 90.0% |
| Hebrew | 91.0% | 94.0% |
| Hindi | 92.0% | 95.0% |
| Hungarian | 90.0% | 92.0% |
| Armenian | 90.0% | 84.0% |
| Indonesian | 96.0% | 93.0% |
| Icelandic | 85.0% | 84.0% |
| Italian | 91.0% | 96.0% |
| Japanese | 97.0% | 97.0% |
| Javanese | 89.0% | 75.0% |
| Georgian | 83.0% | 88.0% |
| Khmer | 84.0% | 81.0% |
| Kannada | 91.0% | 87.0% |
| Korean | 91.0% | 94.0% |
| Latvian | 89.0% | 85.0% |
| Malayalam | 87.0% | 91.0% |
| Mongolian | 88.0% | 84.0% |
| Malay | 93.0% | 92.0% |
| Burmese | 92.0% | 89.0% |
| Norwegian Bokmål | 94.0% | 86.0% |
| Dutch | 92.0% | 90.0% |
| Polish | 93.0% | 92.0% |
| Portuguese | 93.0% | 90.0% |
| Romanian | 91.0% | 94.0% |
| Russian | 97.0% | 94.0% |
| Slovenian | 91.0% | 89.0% |
| Albanian | 94.0% | 76.0% |
| Swedish | 93.0% | 94.0% |
| Swahili | 84.0% | 70.0% |
| Tamil | 89.0% | 92.0% |
| Telugu | 87.0% | 91.0% |
| Thai | 93.0% | 94.0% |
| Filipino | 91.0% | 85.0% |
| Turkish | 90.0% | 92.0% |
| Urdu | 90.0% | 90.0% |
| Vietnamese | 91.0% | 89.0% |
| Chinese (China) | 94.0% | 94.0% |
| Chinese (Taiwan) | 95.0% | 93.0% |
Twenty intents to choose from, a hundred utterances per language.
| Measure | Kai | Jev |
|---|---|---|
| Every option scored (160 or fewer) | ||
| Accuracy, 4 options | 96.8% | 90.5% |
| Accuracy, 16 options | 93.5% | 85.3% |
| Accuracy, 77 options | 90.8% | 84.3% |
| Accuracy, 150 options | 90.8% | 94.0% |
| Shortlisted to 160 by retrieval first | ||
| Accuracy, 1,000 options | 0.5% | refused |
| Accuracy, 10,000 options | 0.0% | refused |
| Accuracy, 100,000 options | 0.0% | refused |
The same cases to both. Jev refuses from 1,000 options. Kai measured on CUDA F32 (dgx); production answers on CPU.
API
One call, every question.
Send a decision program and the evidence it reads.