Claude Opus 5 vs Gemini 3.7 Flash
| model Claude Opus 5 Anthropic | model Gemini 3.7 Flash Google | |
|---|---|---|
| Context | 1M | 1M |
| Input $/M | — | — |
| Output $/M | — | — |
| Knowledge cutoff | 2026-05 | 2026-03 |
| Available via | Anthropic API | Gemini API (AI Studio) |
| SWE-bench Verified (resolved) | 96 | — |
| SWE-bench Pro | 79.2% | — |
| SWE-bench Multilingual | 89.5% | — |
| SWE-bench Multimodal | 59.4% | — |
| DeepSWE | 68.8% | 65.3% |
| FrontierCode (mean@5) | 53.4 | — |
| FrontierCode (score) | — | 43.6 |
| Terminal-Bench (accuracy) | — | 85.8 |
| AutomationBench (accuracy) | — | 30.4 |
| GDP.pdf (accuracy) | — | 34 |
| GDM-MRCR (accuracy) | — | 97 |