Kimi K2.6 (Non-reasoning) — benchmark results
Kimi K2.6 evaluated with reasoning disabled. Provider: Moonshot. Released 2026-04-20. Access: Open.
Unified ELO 1762 ± 32, rank #176 of 1776 rated models, from 35 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| UGI - Writing | 64.19 | Writing Score | 95.7 |
| UGI - Natural Intelligence | 59.46 | NatInt Score | 94.4 |
| AA TAU-2 Bench | 93.86 | Accuracy (%) | 91.1 |
| AA Terminal-Bench Hard | 37.88 | Accuracy (%) | 84.9 |
| AA Omniscience - Software Engineering (SWE) - Rust | 68 | Accuracy (%) | 83.1 |
| AA Omniscience - Software Engineering (SWE) - Go | 40 | Accuracy (%) | 82.9 |
| Artificial Analysis Intelligence Index | 34.58 | Intelligence Index | 82.8 |
| AA Omniscience - Software Engineering (SWE) - PHP | 48 | Accuracy (%) | 82.7 |
| AA Omniscience - Software Engineering (SWE) - Kotlin | 36 | Accuracy (%) | 80.5 |
| AA Omniscience | -9.97 | Score | 79.7 |
| AA Humanity's Last Exam | 18.16 | Accuracy (%) | 77.4 |
| AA Omniscience - Software Engineering (SWE) - C | 57 | Accuracy (%) | 77.2 |
Interactive version: theaggregate.ai/model?slug=kimi-k2-6-non-reasoning · How the rankings work · Data refreshed daily, snapshot 2026-07-22.