Claude Sonnet 5 vs Gemini 3.1 Pro
Claude Sonnet 5 
Gemini 3.1 Pro Verdict
- Gemini 3.1 Pro leads on quality: 64.0 vs 61.4.
- Claude Sonnet 5 (high) is stronger in coding, composite, human preference.
- Gemini 3.1 Pro is stronger in agents & tools, knowledge, long context, multimodal, reasoning, instruction following, maths.
- They cost about the same ($4.00 per 1M blended).
- Gemini 3.1 Pro streams 1.8× faster (108 vs 60 tokens per second).
| Metric | Claude Sonnet 5 (high) | Gemini 3.1 Pro |
|---|---|---|
| BenchLeader Index | 61.4 | 64.0 |
| Agents & tools score | 61.8 | 63.7 |
| Coding score | 61.6 | 59.9 |
| Composite score | 69.5 | 67.4 |
| Human preference score | 66.2 | 59.9 |
| Knowledge score | 62.4 | 66.2 |
| Long context score | 64.8 | 67.6 |
| Multimodal score | 63.1 | 66.2 |
| Reasoning score | 66.8 | 69.8 |
| Instruction following score | – | 69.6 |
| Maths score | – | 60.7 |
| Blended price $/M | $4.00 | $4.50 |
| Output speed | 60 tok/s | 108 tok/s |
| Time to first answer | 10.0 s | 30.5 s |
| Context window | 1M | 1.0M |
| GPQA Diamond | – | 94.1% |
| FrontierMath Tiers 1–3 | – | 59.6% |
| FrontierMath Tier 4 | – | 26.8% |
| OTIS Mock AIME | – | 95.6% |
| Humanity's Last Exam | – | 46.4% |
| Terminal-Bench | – | 80.2% |
| SimpleBench | – | 79.6% |
| SciCode | 48.6% | 58.9% |
| WeirdML | 68.8% | 72.1% |
| APEX-Agents | – | 33.5% |
| ProofBench | – | 26.0% |
| GSO-Bench | – | 22.6% |
| Epoch Capabilities Index | – | 155 |
| LMArena Text | 1462 | 1487 |
| LMArena Hard Prompts | 1490 | 1507 |
| LMArena Coding | 1521 | 1521 |
| LMArena WebDev | 1538 | 1447 |
| LMArena Vision | 1278 | 1295 |
| LMArena Agent | 6.3 | -5.3 |
| AA Intelligence Index | 32.0 | 30.4 |
| IFBench | – | 77.1% |
| AA-LCR | 76.7% | 82.0% |
| MMMU-Pro | – | 82.4% |
| AA-Omniscience | -3.7 | 31.9 |
| Terminal-Bench Hard | – | 53.8% |
| GPQA Diamond (AA) | – | 94.1% |
| Humanity's Last Exam (AA) | 35.7% | 47.0% |
| SciCode (AA) | 54.3% | 58.7% |
| τ²-Bench Telecom (AA) | – | 95.6% |
| 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 | – | 35.3% |
| TutorBench | – | 53.0% |
| EQ-Bench 4 | – | 1142 |
| ARC-AGI-1 | – | 98.0% |
| ARC-AGI-2 | – | 77.1% |
| ARC-AGI-3 | – | 0.4% |
Data as of 2026-09-10. Best configuration of each model; every score links to its source on the model pages.
Claude Sonnet 5 vs Gemini 3.1 Pro: questions
- Is Claude Sonnet 5 better than Gemini 3.1 Pro?
- Gemini 3.1 Pro leads on quality: 64.0 vs 61.4. The BenchLeader Index combines every independent quality benchmark; Gemini 3.1 Pro is ahead overall as of 2026-09-10, but check the category scores for your use.
- Is Claude Sonnet 5 better than Gemini 3.1 Pro for coding?
- Claude Sonnet 5 scores higher in coding (62 vs 60 on the category index, where 50 is average).
- Is Claude Sonnet 5 better than Gemini 3.1 Pro for agentic tasks?
- Gemini 3.1 Pro scores higher in agentic tasks (64 vs 62 on the category index, where 50 is average).
- Which is cheaper, Claude Sonnet 5 or Gemini 3.1 Pro?
- Claude Sonnet 5 is cheaper: $4.00 against $4.50 per million tokens, blended at three input tokens per output token.
- Which is faster, Claude Sonnet 5 or Gemini 3.1 Pro?
- Gemini 3.1 Pro streams faster: 108 against 60 output tokens per second.
- Which has the larger context window?
- Gemini 3.1 Pro accepts more context: 1.0M against 1M tokens.