SIDE BY SIDE
AI Model Comparison: Side-by-Side Prices and Capabilities
Compare uses, prices, speed evidence, and version details for up to three models. Language, video, decision models and historical evaluations are shown separately.
Historical benchmark snapshot
| Evaluation metric | Gemini 3.1 Pro ↗High · Google | Claude Opus 4.6 ↗Max · Anthropic |
|---|---|---|
| SWE-bench VerifiedResolved % | 80.6 | 80.8 |
| GPQA DiamondPass@1 % | 94.3 | 91.3 |
| HMMT 2026 FebPass@1 % | 94.7 | 96.2 |
| MMLU-ProEM % | 91.0 | 89.1 |
| LiveCodeBenchPass@1 % | 91.7 | 88.8 |
| Terminal-Bench 2.0Accuracy % | 68.5 | 65.4 |
| SWE-bench ProResolved % | 54.2 | 57.3 |
| Humanity’s Last ExamPass@1 % | 44.4 | 40.0 |
| BrowseCompPass@1 % | 85.9 | 83.7 |
| MRCR 1MMMR % | 76.3 | 92.9 |
| Model weights | Closed weights | Closed weights |
Source: DeepSeek V4 preview model card. Checked 2026-09-24. Highest reported values appear in green, but reasoning budgets and tool frameworks may differ. These scores are not an evaluation of newer models.