Compare / Claude Sonnet 5 vs Gemini 3.5 Flash
Claude Sonnet 5 vs Gemini 3.5 Flash
23 results from 6 sources that measured both models. Each row is in the source's own units; there is no overall winner.
| Benchmark | Metric | Claude Sonnet 5 | Gemini 3.5 Flash |
|---|---|---|---|
| APEX-Agentsreported | Tasks passed% of tasks | 54.5% | 27.5% |
| Arena (formerly LMArena)reported | OverallArena rating | 1,476 | 1,461 |
| Arena (formerly LMArena)reported | OverallArena rating | 1,452 | 1,476 |
| Arena (formerly LMArena)reported | Business, management and financeArena rating | 1,465 | 1,468 |
| Arena (formerly LMArena)reported | Creative writingArena rating | 1,435 | 1,466 |
| Arena (formerly LMArena)reported | Expert promptsArena rating | 1,512 | 1,491 |
| Arena (formerly LMArena)reported | Instruction followingArena rating | 1,467 | 1,465 |
| Arena (formerly LMArena)reported | OverallArena rating | 1,462 | 1,477 |
| Arena (formerly LMArena)reported | Writing, literature and languageArena rating | 1,452 | 1,472 |
| Artificial Analysisreported | Artificial Analysis Coding Indexindex score | 71.5 | 70.1 |
| Artificial Analysisreported | Artificial Analysis Intelligence Indexindex score | 38.2 | 33.6 |
| Artificial Analysisreported | GPQA Diamond% of questions | 91.1% | 92.2% |
| Artificial Analysisreported | Humanity's Last Exam% of questions | 41.3% | 42.7% |
| Artificial Analysisreported | Long-context reasoning (AA-LCR)% of questions | 82.0% | 74.3% |
| Artificial Analysisreported | SciCode% of problems | 54.3% | 53.9% |
| Artificial Analysisreported | Terminal-Bench 2.1% of tasks | 80.5% | 78.7% |
| Artificial Analysisreported | Terminal-Bench 4.0% of tasks | 14.1% | 6.6% |
| Artificial Analysisreported | Τ-bench banking% of tasks | 37.3% | 32.2% |
| SimpleQA Verified (Epoch AI)reported | Correct answers% of questions | 33.7% | 66.2% |
| UGI Leaderboardreported | Requested-length error% off the requested word count | 3.0% | 9.0% |
| UGI Leaderboardreported | Style adherencescore from 0 to 1 | 0.38 | 0.35 |
| UGI Leaderboardreported | Writing scorescore out of 100 | 62.7 | 72.5 |
| Vending-Bench 2reported | Money after a yearUS dollars | $6,377.70 | $5,396.42 |
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.