Kimi K3 — Benchmarks
Benchmark scores for Kimi K3 aggregated from public leaderboards, with how it ranks among open models. See hardware requirements for what you need to run it.
Overall rank: #8 of 123 open modelscomposite 70.6/100 across 11 benchmarks in 4 categories · methodology
Coding
| Benchmark | Score | Open rank | All models |
|---|---|---|---|
| LiveBench Coding | 81.5 | #1 / 16 | #11 / 54 |
| SciCode | 58.7% | #1 / 50 | #5 / 155 |
Knowledge
| Benchmark | Score | Open rank | All models |
|---|---|---|---|
| SimpleQA | 50.6% | #2 / 15 | #19 / 78 |
Math
| Benchmark | Score | Open rank | All models |
|---|---|---|---|
| LiveBench Math | 84.4 | #10 / 16 | #41 / 54 |
| AIME 2024/2025 | 97.2% | #2 / 74 | #23 / 270 |
| ProofBench | 87.0% | #1 / 21 | #5 / 64 |
Reasoning
| Benchmark | Score | Open rank | All models |
|---|---|---|---|
| LiveBench Reasoning | 90.7 | #1 / 16 | #5 / 54 |
| GPQA Diamond | 93.1% | #1 / 83 | #16 / 291 |
| ARC-AGI | 94.5% | #1 / 16 | #27 / 200 |
| SimpleBench | 60.7% | #2 / 23 | #29 / 101 |
| CritPt | 23.4% | #1 / 50 | #27 / 167 |
Scores aggregated from public benchmark sources (each linked from the benchmark pages). llmrun does not run these benchmarks.