Claude Opus 5 vs Gemini 3.1 Pro
Claude Opus 5 
Gemini 3.1 Pro Verdict
- Claude Opus 5 leads on quality: 70.0 vs 63.9.
- Claude Opus 5 is stronger in agents & tools, coding, composite, human preference, knowledge, maths, multimodal, reasoning.
- Gemini 3.1 Pro is stronger in long context, instruction following.
- Gemini 3.1 Pro is 2.2× cheaper ($4.50 vs $10.00 per 1M blended).
- Gemini 3.1 Pro streams 2.1× faster (110 vs 52 tokens per second).
| Metric | Claude Opus 5 | Gemini 3.1 Pro |
|---|---|---|
| BenchLeader Index | 70.0 | 63.9 |
| Agents & tools score | 71.6 | 63.7 |
| Coding score | 74.4 | 59.8 |
| Composite score | 92.9 | 67.2 |
| Human preference score | 78.2 | 59.6 |
| Knowledge score | 70.5 | 66.1 |
| Long context score | 66.2 | 67.6 |
| Maths score | 67.1 | 60.7 |
| Multimodal score | 69.4 | 66.1 |
| Reasoning score | 71.1 | 69.5 |
| Instruction following score | – | 69.5 |
| Blended price $/M | $10.00 | $4.50 |
| Output speed | 52 tok/s | 110 tok/s |
| Time to first answer | 81.6 s | 30.5 s |
| Context window | 1M | 1.0M |
| GPQA Diamond | 92.9% | 94.1% |
| FrontierMath Tiers 1–3 | – | 59.6% |
| FrontierMath Tier 4 | – | 26.8% |
| OTIS Mock AIME | 97.8% | 95.6% |
| Humanity's Last Exam | – | 46.4% |
| Terminal-Bench | – | 80.2% |
| SimpleBench | 80.6% | 79.6% |
| SciCode | – | 58.9% |
| WeirdML | 86.3% | 72.1% |
| APEX-Agents | – | 33.5% |
| ProofBench | – | 26.0% |
| GSO-Bench | – | 22.6% |
| Epoch Capabilities Index | 162.6 | 155 |
| LMArena Text | – | 1487 |
| LMArena Hard Prompts | – | 1507 |
| LMArena Coding | – | 1521 |
| LMArena WebDev | – | 1447 |
| LMArena Vision | – | 1295 |
| LMArena Agent | – | -5.3 |
| AA Intelligence Index | 50.7 | 30.4 |
| IFBench | – | 77.1% |
| AA-LCR | 79.3% | 82.0% |
| MMMU-Pro | 84.7% | 82.4% |
| AA-Omniscience | 37.1 | 31.9 |
| Terminal-Bench Hard | – | 53.8% |
| GPQA Diamond (AA) | 93.2% | 94.1% |
| Humanity's Last Exam (AA) | 54.9% | 47.0% |
| SciCode (AA) | 56.4% | 58.7% |
| τ²-Bench Telecom (AA) | – | 95.6% |
| LiveCodeBench | 89.0% | – |
| MMLU-Pro | 91.6% | – |
| IOI | 91.7% | – |
| LegalBench | 87.0% | – |
| CorpFin | 73.2% | – |
| TaxEval | 75.1% | – |
| Terminal-Bench 2.1 (Vals) | 84.6% | – |
| SWE-bench (Vals) | 97.0% | – |
| GPQA Diamond (Vals) | 93.4% | – |
| Vals Index | 67.2 | – |
| AIME 2026 | – | 98.3% |
| HMMT February 2026 | – | 94.7% |
| MathArena Apex | – | 60.9% |
| MultiChallenge | – | 71.4% |
| PRBench Finance | – | 41.9% |
| PRBench Legal | – | 44.0% |
| MultiNRC | – | 64.7% |
| HiL-Bench | 57.0% | 35.3% |
| TutorBench | – | 53.0% |
| EQ-Bench 4 | 1385 | 1142 |
| ARC-AGI-1 | – | 98.0% |
| ARC-AGI-2 | – | 77.1% |
| ARC-AGI-3 | – | 0.4% |
Data as of 2026-09-09. Best configuration of each model; every score links to its source on the model pages.