Claude Sonnet 4.6 vs GPT-5.6 Sol
Claude Sonnet 4.6 
GPT-5.6 Sol Verdict
- GPT-5.6 Sol (high) leads on quality: 68.2 vs 60.4.
- Claude Sonnet 4.6 is stronger in human preference, maths, reasoning.
- GPT-5.6 Sol (high) is stronger in agents & tools, coding, composite, instruction following, knowledge, long context, multimodal.
- Claude Sonnet 4.6 is 1.3× cheaper ($6.00 vs $8.00 per 1M blended).
- GPT-5.6 Sol (high) streams 1.4× faster (59 vs 42 tokens per second).
| Metric | Claude Sonnet 4.6 | GPT-5.6 Sol (high) |
|---|---|---|
| BenchLeader Index | 60.4 | 68.2 |
| Agents & tools score | 54.6 | 87.9 |
| Coding score | 57.1 | 72.4 |
| Composite score | 67.5 | 82.7 |
| Human preference score | 62.0 | – |
| Instruction following score | 56.9 | 67.8 |
| Knowledge score | 62.3 | 73.8 |
| Long context score | 66.6 | 67.4 |
| Maths score | 62.7 | – |
| Multimodal score | 60.6 | 66.4 |
| Reasoning score | 65.2 | 63.3 |
| Blended price $/M | $6.00 | $8.00 |
| Output speed | 42 tok/s | 59 tok/s |
| Time to first answer | 2.2 s | 31.0 s |
| Context window | 1M | 1M |
| GPQA Diamond | 87.4% | – |
| OTIS Mock AIME | 85.8% | – |
| SWE-bench Verified (Epoch) | 75.2% | – |
| Terminal-Bench | 53.4% | – |
| SciCode | – | 56.9% |
| WeirdML | – | 88.8% |
| FrontierCode | 24.3% | – |
| Epoch Capabilities Index | 152.3 | – |
| LMArena Text | 1472 | – |
| LMArena Hard Prompts | 1504 | – |
| LMArena Coding | 1528 | – |
| LMArena WebDev | 1521 | – |
| LMArena Vision | 1282 | – |
| LMArena Agent | -1 | – |
| AA Intelligence Index | 30.4 | 42.5 |
| IFBench | 56.6% | 69.2% |
| AA-LCR | 80.0% | 81.7% |
| MMMU-Pro | 73.3% | 81.8% |
| AA-Omniscience | 12.2 | 20.4 |
| Terminal-Bench Hard | 53.0% | 62.1% |
| GPQA Diamond (AA) | 87.5% | 92.8% |
| Humanity's Last Exam (AA) | 33.6% | 46.0% |
| SciCode (AA) | 50.1% | 57.8% |
| τ²-Bench Telecom (AA) | 79.5% | 83.3% |
| AIME (Vals) | 92.3% | – |
| LiveCodeBench | 82.1% | – |
| MMLU-Pro | 87.3% | – |
| LegalBench | 82.1% | – |
| CorpFin | 65.3% | – |
| TaxEval | 77.1% | – |
| MedQA | 92.1% | – |
| Terminal-Bench 2.1 (Vals) | 57.3% | – |
| SWE-bench (Vals) | 77.4% | – |
| GPQA Diamond (Vals) | 85.6% | – |
| Vals Index | 50.6 | – |
| MCP Atlas | 69.5% | – |
| EnigmaEval | – | 37.1% |
| EQ-Bench 4 | 1207 | – |
| ARC-AGI-1 | – | 97.0% |
| ARC-AGI-2 | – | 85.4% |
| ARC-AGI-3 | – | 2.2% |
Data as of 2026-09-10. Best configuration of each model; every score links to its source on the model pages.
Claude Sonnet 4.6 vs GPT-5.6 Sol: questions
- Is Claude Sonnet 4.6 better than GPT-5.6 Sol?
- GPT-5.6 Sol (high) leads on quality: 68.2 vs 60.4. The BenchLeader Index combines every independent quality benchmark; GPT-5.6 Sol (high) is ahead overall as of 2026-09-10, but check the category scores for your use.
- Is Claude Sonnet 4.6 better than GPT-5.6 Sol for coding?
- GPT-5.6 Sol scores higher in coding (72 vs 57 on the category index, where 50 is average).
- Is Claude Sonnet 4.6 better than GPT-5.6 Sol for agentic tasks?
- GPT-5.6 Sol scores higher in agentic tasks (88 vs 55 on the category index, where 50 is average).
- Which is cheaper, Claude Sonnet 4.6 or GPT-5.6 Sol?
- Claude Sonnet 4.6 is cheaper: $6.00 against $8.00 per million tokens, blended at three input tokens per output token.
- Which is faster, Claude Sonnet 4.6 or GPT-5.6 Sol?
- GPT-5.6 Sol streams faster: 59 against 42 output tokens per second.
- Which has the larger context window?
- Both accept 1M tokens of context.