Kimi K2.7 Code model card (Hugging Face) 12 Jun 2026
- Kimi Code Bench v2 62
- Program Bench 53.6
- MLS Bench Lite 35.1
- Kimi Claw 24/7 Bench 46.9
- MCP Atlas 76
- MCP Mark Verified 81.1
Kimi K2.7 Code
Kimi K2.7 Code ranks first on LiveClawBench.
36
Benchmarks with a score
1
Ranked first, out of five or more models
2
In the top three
Moonshot AI Jun 2026
Kimi K2.7 Code
Open weights Mid-size
150.2
#61 of 271
Epoch Capabilities Index
- DeepSWE 30.5% #26 /28
- CritPt 10% #59 /160
- APEX-Agents 37.6% #46 /56
Where it ranks
Its best score on each benchmark, and where that ranks among every model with a score there. We rank each model once, at its best setting.
Knowledge and language
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| SimpleQA Verified | 36.5% | #53 of 84 | Epoch AI |
Math and science
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| OTIS Mock AIME 2024-2025 | 95.6% | #43 of 202 | Epoch AI | |
| GPQA Diamond | 87.9% | #57 of 219 | Epoch AI | |
| CritPt | 10% | #59 of 160 | Epoch AI | |
| SimpleBench | 57.9% | #42 of 100 | Official leaderboard | |
| FrontierMath Tiers 1-3 v2 | 54% | #51 of 89 | Epoch AI | |
| Surface Evolver Bench | 48.8% | #16 of 26 | Epoch AI | |
| FrontierMath Tier 4 v2 | 12.2% | #59 of 71 | Epoch AI |
Coding
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| WeirdML | 54.1% | #41 of 112 | Official leaderboard | |
| WebDev Arena | 1,473 | #50 of 105 | Epoch AI | |
| Code Arena | 1,474 | #62 of 124 | Official leaderboard | |
| AndroidBench | 70.4% | #16 of 32 | Official leaderboard | |
| ALE-Bench | 886 | #65 of 121 | Epoch AI | |
| Vibe Code Bench | 47.2% | #65 of 115 | Official leaderboard | |
| ProgramBench | 53.6% | #12 of 20 | Model card | |
| Code Migration | 25.4% | #48 of 78 | Official leaderboard | |
| Image-to-WebDev Arena | 1,513 | #36 of 55 | Official leaderboard | |
| FrontierCode | 30.1% | #32 of 46 | Epoch AI | |
| MLS-Bench | 35.1% | #11 of 15 | Official leaderboard | |
| EvoCode-Bench v2 | 7.80% | #10 of 13 | Official leaderboard | |
| GBAEval | 0.85% | #19 of 21 | Official leaderboard | |
| DeepSWE | 30.5% | #26 of 28 | Official leaderboard | |
| Next.js Agent Evals | 74% | #14 of 14 | Official leaderboard | |
| Kimi Code Bench v2 | 62% | #3 of 3 | Model card |
Agents and tools
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| LiveClawBench | 76% | #1 of 17 | Official leaderboard | |
| MCPMark | 81.1% | #2 of 10 | Model card | |
| Pencil Puzzle Bench | 16.7% | #16 of 53 | Official leaderboard | |
| PinchBench 2.0 | 76.1% | #20 of 57 | Official leaderboard | |
| MCP Atlas | 76% | #18 of 40 | Model card | |
| Long-Horizon-Terminal-Bench (LHTB) | 36.7% | #14 of 25 | Official leaderboard | |
| Toolathlon-Verified | 58% | #14 of 25 | Official leaderboard | |
| Chess Puzzles | 14% | #71 of 83 | Epoch AI |
Professional work
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| Vending-Bench 2 | $5,083 | #35 of 66 | Official leaderboard | |
| APEX-Agents | 37.6% | #46 of 56 | Epoch AI | |
| Kimi Claw 24/7 Bench | 46.9% | #1 of 1 | Model card |
Safety and security
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| DrugDiscoveryBench | 35.3% | #13 of 19 | Official leaderboard |
Moonshot AI's models over time
Moonshot AI's models in color, against every other model on the Epoch Capabilities Index, by release date.
Its launch post
The benchmarks the lab chose to headline, with the scores it gave.