Claude Sonnet 5 vs GPT-6.1 Sol
Claude Sonnet 5 
GPT-6.1 Sol Verdict
- GPT-6.1 Sol (max) leads on quality: 69.3 vs 60.5.
- GPT-6.1 Sol (max) is stronger in agents & tools, coding, composite, human preference, knowledge, long context, maths, multimodal, reasoning.
- They cost about the same ($4.00 per 1M blended).
| Metric | Claude Sonnet 5 (high) | GPT-6.1 Sol (max) |
|---|---|---|
| BenchLeader Index | 60.5 | 69.3 |
| Agents & tools score | 57.9 | 63.9 |
| Coding score | 61.1 | 72.4 |
| Composite score | 65.4 | 79.0 |
| Human preference score | 65.3 | 68.1 |
| Knowledge score | 60.0 | 78.8 |
| Long context score | 63.5 | 66.8 |
| Maths score | 65.8 | 72.5 |
| Multimodal score | 61.3 | 66.2 |
| Reasoning score | 66.3 | 76.4 |
| Blended price $/M | $4.00 | $4.00 |
| Output speed | 62 tok/s | 56 tok/s |
| Time to first answer | 12.3 s | 326.9 s |
| Context window | 1M | 1.1M |
| GPQA Diamond | – | 95.4% |
| FrontierMath Tiers 1–3 | – | 93.7% |
| FrontierMath Tier 4 | – | 100.0% |
| OTIS Mock AIME | – | 100.0% |
| SimpleQA Verified | – | 73.9% |
| Terminal-Bench | – | 58.2% |
| SciCode | 54.3% | 54.2% |
| WeirdML | 68.8% | – |
| APEX-Agents | – | 60.0% |
| LMArena Text | 1461 | 1484 |
| LMArena Hard Prompts | 1490 | 1507 |
| LMArena Coding | 1519 | 1545 |
| LMArena WebDev | 1541 | 1755 |
| LMArena Vision | 1274 | 1288 |
| LMArena Agent | 4.9 | 11.7 |
| LiveBench | – | 81.6% |
| LiveBench Reasoning | – | 92.6% |
| LiveBench Coding | – | 80.4% |
| LiveBench Agentic Coding | – | 54.5% |
| LiveBench Mathematics | – | 96.8% |
| LiveBench Data Analysis | – | 82.7% |
| LiveBench Language | – | 90.1% |
| LiveBench Instruction Following | – | 74.2% |
| AA Intelligence Index v4.3.2 | 31.7 | 51.8 |
| AA-LCR | 76.7% | 83.0% |
| MMMU-Pro | – | 86.0% |
| AA-Omniscience | -3.7 | 41.5 |
| Humanity's Last Exam (AA) | 35.7% | 52.9% |
| SciCode (AA) | 54.3% | 54.2% |
| IOI | – | 96.9% |
| Vals Index | – | 61.1 |
| ARC-AGI-1 | – | 96.5% |
| ARC-AGI-2 | – | 94.2% |
| ARC-AGI-3 | – | 96.2% |
| CritPt | 15.1% | 31.7% |
| GDPval-AA v2.1 | 37.9% | 53.8% |
| Analyst Agent (AA) | – | 50.0% |
| BioMysteryBench | – | 79.6% |
| Code Migration | – | 65.1% |
| CyberBench | – | 39.3% |
| Excel Modeling Benchmark | – | 70.8% |
| Finance Agent v2 | – | 52.0% |
| Harvey's Legal Agent Benchmark | – | 5.4% |
| Legal Research Bench | – | 38.5% |
| MedCode | – | 48.8% |
| MedScribe | – | 86.5% |
| MysteryMechanism | – | 46.4% |
| Public Benefits Bench | – | 59.3% |
| SAGE | – | 46.5% |
| SREBench | – | 50.8% |
| Tax Agent Bench | – | 62.3% |
| Terminal-Bench 4.0 (Vals) | – | 55.0% |
| Vibe Code Bench v1.1 | – | 88.9% |
| LMArena Maths | 1473 | 1488 |
| LMArena Creative Writing | 1435 | 1462 |
| LMArena Instruction Following | 1466 | 1488 |
| LMArena Multi-turn | 1473 | 1487 |
| LMArena Longer Queries | 1481 | 1496 |
| LMArena Document | 1476 | – |
| Chess Puzzles | – | 61.0% |
| EBR-bench | – | 54.3% |
| Mystery Game Puzzles | – | 80.0% |
| DeepSWE v1.1 | 48.2% | – |
| LMCA | 50.0% | – |
| DTBench | 84.5% | – |
| CursorBench | 30.8% | – |
| ALE-Bench | 1463.1 | – |
| Terminal-Bench 4.0 (AA) | 5.0% | 56.1% |
| AutomationBench | – | 64.9% |
| GDP.pdf | – | 31.0% |
| MLCR | – | 33.9% |
| Harvey LAB | – | 6.9% |
| AA-Omniscience: accuracy | 37.4% | 62.1% |
| AA-Omniscience: non-hallucination | 34.3% | 45.7% |
| AA-Briefcase v1.1 | – | 1557 |
Data as of 2026-10-11. Best configuration of each model; every score links to its source on the model pages.
Claude Sonnet 5 vs GPT-6.1 Sol: questions
- Is Claude Sonnet 5 better than GPT-6.1 Sol?
- GPT-6.1 Sol (max) leads on quality: 69.3 vs 60.5. The BenchLeader Index combines every independent quality benchmark; GPT-6.1 Sol (max) is ahead overall as of 2026-10-11, but check the category scores for your use.
- Is Claude Sonnet 5 better than GPT-6.1 Sol for coding?
- GPT-6.1 Sol scores higher in coding (72 vs 61 on the category index, where 50 is average).
- Is Claude Sonnet 5 better than GPT-6.1 Sol for agentic tasks?
- GPT-6.1 Sol scores higher in agentic tasks (64 vs 58 on the category index, where 50 is average).
- Which is cheaper, Claude Sonnet 5 or GPT-6.1 Sol?
- Claude Sonnet 5 is cheaper: $4.00 against $4.00 per million tokens, blended at three input tokens per output token.
- Which is faster, Claude Sonnet 5 or GPT-6.1 Sol?
- Claude Sonnet 5 streams faster: 62 against 56 output tokens per second.
- Which has the larger context window?
- GPT-6.1 Sol accepts more context: 1.1M against 1M tokens.