Kimi K2 Base
Moonshot
Highlights
Top benchmark results for moonshotai/kimi-k2-base-2025-07-11.
0.93#1
0.78#2
0.80#1
0.48#59
0.92#1
0.26#6
0.40#3
0.88#6
0.90#1
0.69#9
0.35#6
0.45#4
0.85#1
Benchmark table
Detailed scores across tracked benchmarks.
| Benchmark | Category | Top Score | Info | Self Reported | Source |
|---|---|---|---|---|---|
| C-Eval | - | 0.93 | EM | Yes | Source |
| CSimpleQA | - | 0.78 | Correct | Yes | Source |
| EvalPlus | - | 0.80 | Pass@1 | Yes | Source |
| GPQA Diamond | general-knowledge | 0.48 | Avg@8 | Yes | Source |
| GSM8K | - | 0.92 | EM | Yes | Source |
| LiveCodeBench V6 | - | 0.26 | Pass@1 | Yes | Source |
| MATH | - | 0.40 | EM | Yes | Source |
| MMLU | - | 0.88 | EM | Yes | Source |
| MMLU Redux 2.0 | - | 0.90 | EM | Yes | Source |
| MMLU-Pro | - | 0.69 | EM | Yes | Source |
| SimpleQA | - | 0.35 | Correct | Yes | Source |
| SuperGPQA | - | 0.45 | EM | Yes | Source |
| TriviaQA | - | 0.85 | EM | Yes | Source |
Benchmark comparisons
Use the selector to switch benchmarks and see how this model stacks up against its closest competitors.
GSM8K
Compare this model with the leading peers for the selected benchmark.
Benchmark
0.92
Rank #1/3
3 models
Showing 3 models around the selected model (out of 3 total).