Kimi K2.6 (Thinking): benchmark results
Moonshot Kimi K2.6 evaluated with thinking enabled. Provider: Moonshot. Released 2026-04-20. Access: Open.
Aggregate Rating 1724 ± 7, rank #293 of 2168 rated models, from 269 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| LLM Stats (MMLU-Redux) | 95.3 | Score (%) | 100 |
| Qwen3.7 Launch - Humanity's Last Exam (with tools) | 54 | Score (%) | 100 |
| SuperCLUE-Writing - Content Creativity | 89.39 | Score | 100 |
| AA TAU-2 Bench | 95.91 | Accuracy (%) | 96.2 |
| PatternEval - Response-Pattern Failure Rate (Thinking) | 0.99 | Bad-pattern trigger rate (%, lower is better; share of final | 95.8 |
| VitaBench 2.0 | 48.1 | Avg@4 Full Context (imported) | 95.5 |
| AA IFBench | 75.99 | Accuracy (%) | 94.9 |
| UGI - Natural Intelligence | 65.25 | NatInt Score | 94.5 |
| RuVerBench - Deep Research | 92.2 | Balanced Accuracy (%) | 94.1 |
| UGI - Writing | 64.4 | Writing Score | 94.1 |
| LLM Stats (PolyMATH) | 82.7 | Score (%) | 92.3 |
| SuperCLUE-Writing - Content Style | 90.96 | Score | 92.3 |
Interactive version: theaggregate.ai/model?slug=kimi-k2-6-thinking · How It Works · Data refreshed daily, snapshot 2026-10-20.