Compare / Gemini 3.6 Flash vs Kimi K2.7 Code
Gemini 3.6 Flash vs Kimi K2.7 Code
11 results from 3 sources that measured both models. Each row is in the source's own units; there is no overall winner.
| Benchmark | Metric | Gemini 3.6 Flash | Kimi K2.7 Code |
|---|---|---|---|
| APEX-Agentsreported | Tasks passed% of tasks | 46.9% | 37.6% |
| Artificial Analysisreported | Artificial Analysis Coding Indexindex score | 69.2 | 60.8 |
| Artificial Analysisreported | Artificial Analysis Intelligence Indexindex score | 34.0 | 25.8 |
| Artificial Analysisreported | GPQA Diamond% of questions | 92.8% | 89.6% |
| Artificial Analysisreported | Humanity's Last Exam% of questions | 40.8% | 35.0% |
| Artificial Analysisreported | Long-context reasoning (AA-LCR)% of questions | 80.0% | 79.3% |
| Artificial Analysisreported | SciCode% of problems | 53.4% | 47.8% |
| Artificial Analysisreported | Terminal-Bench 2.1% of tasks | 77.5% | 67.4% |
| Artificial Analysisreported | Terminal-Bench 4.0% of tasks | 7.1% | 1.0% |
| Artificial Analysisreported | Τ-bench banking% of tasks | 29.9% | 20.2% |
| SimpleQA Verified (Epoch AI)reported | Correct answers% of questions | 66.2% | 36.5% |
Bold green marks the better value on that metric. Where a source reports ranges that overlap, the difference may not be meaningful; see the benchmark page for ranges.