Moonshot AI
Kimi K2 (Jul 2025) benchmark results
9 results on 4 benchmarks from 4 sources.
Open weights Released: Jul 12, 2025 Updated Oct 22, 2026
Also listed as: kimi-k2-0711-preview · Kimi-K2-Instruct · Kimi K2 Instruct · moonshot/kimi-k2-0711-preview · Kimi K2 · OpenHands + Kimi K2 · kimi-k2-instruct · CodeSweep - SWE-agent - Kimi K2 Instruct
All benchmark results
| Benchmarks | Score | Rank | Setting | Source type | Date |
|---|---|---|---|---|---|
| LMArena Text | 1418 | 133 | Overall, style control on; 27148 votes; 95% CI 1413–1423“kimi-k2-0711-preview” | Official leaderboard | Oct 8, 2026 |
| MMLU-Pro | 81.0% | 59 | Overall accuracy; data source: Self-Reported; 1000B“Kimi-K2-Instruct” | Vendor-reported | Mar 11, 2026 |
| SWE-bench Verified | 65.4% | 37 | agent: OpenHands“moonshot/kimi-k2-0711-preview” | Official leaderboard | Jul 16, 2025 |
| SWE-bench Verified | 53.4% | 64 | agent: CodeSweep - SWE-agent“kimi-k2-instruct” | Official leaderboard | Aug 4, 2025 |
| SWE-bench Verified | 43.8% | 76 | agent: mini-SWE-agent“Kimi-K2-Instruct” | Official leaderboard | Aug 7, 2025 |
| Terminal-Bench 2.0 | 27.8% | 136 | Terminal-Bench 2.0; agent: Terminus 2 | Official leaderboard via Epoch AI | Nov 1, 2025 |
| Terminal-Bench 2.0 | 27.8% | 137 | Terminal-Bench 2.0; agent: Terminus 2 | Official leaderboard via Epoch AI | Jul 11, 2025 |
| Terminal-Bench 2.0 | 26.7% | 143 | Terminal-Bench 2.0; agent: OpenHands | Official leaderboard via Epoch AI | Jul 11, 2025 |
| Terminal-Bench 2.0 | 26.7% | 144 | Terminal-Bench 2.0; agent: OpenHands | Official leaderboard via Epoch AI | Nov 2, 2025 |
FAQ
Where do these Kimi K2 (Jul 2025) scores come from?
From LMArena leaderboard dataset, MMLU-Pro leaderboard (TIGER-Lab), SWE-bench leaderboards (swebench.com), and Epoch AI Benchmarking Hub. Each row links to the original leaderboard, dataset or report, and vendor-reported numbers are labelled.