Moonshot AI
Kimi K2.6 benchmark results
7 results on 7 benchmarks from 2 sources.
Also listed as: kimi-k2.6 · Kimi 2.6 (enabled)
All benchmark results
| Benchmarks | Score | Rank | Setting | Source type | Date |
|---|---|---|---|---|---|
| LMArena Text | 1461 | 58 | Overall, style control on; 40066 votes; 95% CI 1456–1465“kimi-k2.6” | Official leaderboard | Oct 8, 2026 |
| LMArena WebDev | 1509 | 54 | Overall; 10187 votes; 95% CI 1502–1516“kimi-k2.6” | Official leaderboard | Oct 8, 2026 |
| LMArena Vision | 1265 | 38 | Overall, style control on; 16387 votes; 95% CI 1258–1272“kimi-k2.6” | Official leaderboard | Oct 2, 2026 |
| GPQA Diamond | 90.8% | 34 | Epoch AI run, accuracy | Independent run | May 1, 2026 |
| SWE-bench Verified | 76.7% | 9 | Epoch AI run (Epoch scaffold) | Independent run | May 8, 2026 |
| AIME (OTIS Mock AIME 2024–2025) | 96.1% | 40 | Epoch AI run, OTIS Mock AIME 2024–2025 | Independent run | May 2, 2026 |
| OSWorld 2.0 | 4.6% | 12 | Binary accuracy; 500 steps; reasoning: enabled | Official leaderboard via Epoch AI | Apr 20, 2026 |
Related on this site
FAQ
Where do these Kimi K2.6 scores come from?
From Epoch AI Benchmarking Hub and LMArena leaderboard dataset. Each row links to the original leaderboard, dataset or report, and vendor-reported numbers are labelled.