Kimi K2 Turbo (Preview): benchmark results
Provider: Moonshot. Released 2025-08-01.
Aggregate Rating 1636 ± 28, rank #721 of 10916 rated models, from 14 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| KyrgyzLLM-Bench - KyrgyzMMLU | 56.78 | Accuracy (%, zero-shot) | 72 |
| KyrgyzLLM-Bench - KyrgyzRC | 88 | Accuracy (%, zero-shot) | 72 |
| AIDABench - Data Visualization | 67.25 | Score (%) on the 116 data-visualization tasks, best of three | 30 |
| Design Arena (Game Dev) | 1082 | Elo | 26.8 |
| Design Arena (Website) | 1087 | Elo | 22.3 |
| AIDABench | 55.56 | Score (%) over all 603 tasks (category-weighted mean of the | 20 |
| AIDABench (Pass@1) - File Generation | 27.59 | Score (%) on the 261 file-generation tasks, a single run (pa | 20 |
| AIDABench - Easy | 74.31 | Score (%) on the 178 easy tasks (at most 6 expert operation | 20 |
| AIDABench - File Generation | 44.06 | Score (%) on the 261 file-generation tasks, best of three ru | 20 |
| AIDABench - Hard | 29.38 | Score (%) on the 127 hard tasks (13 or more expert operation | 20 |
| AIDABench - Medium | 55.51 | Score (%) on the 298 medium tasks (7 to 12 expert operation | 20 |
| AIDABench - QA | 62.83 | Score (%) on the 226 question-answering tasks, best of three | 20 |
Interactive version: theaggregate.ai/model?slug=kimi-k2-turbo-preview · How It Works · Data refreshed daily, snapshot 2026-11-08.