Model comparison
Claude Sonnet 5 vs GPT-6 Sol
GPT-6 Sol is the stronger model overall, scoring 61.8 to 54.6 on the Noometry Index.
Last verified . 39 shared benchmarks.
Summary
- They share 39 benchmarks with published results for both. Claude Sonnet 5 scores higher in 4 categories and GPT-6 Sol in 6 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where GPT-6 Sol leads 74.0 to 49.1.
- The biggest single-benchmark swing is FrontierMath Tier 4: 29.3% for Claude Sonnet 5 and 90% for GPT-6 Sol.
- Both cost about the same: $2 input and $10 output per million tokens.
- GPT-6 Sol accepts more context: 1.05M tokens versus 1M.
Side by side
| Claude Sonnet 5 | GPT-6 Sol | |
|---|---|---|
| Provider | Anthropic | OpenAI |
| Noometry Index | 54.6 | 61.8 |
| Released | 2026-06-29 | 2026-09-22 |
| Weights | Proprietary | Proprietary |
| Context window | 1M | 1.05M |
| Max output | 128K | 128K |
| Input $ / M tokens | $2 | $2 |
| Output $ / M tokens | $10 | $10 |
| Results tracked | 51 | 45 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-6 Sol leads
Claude Sonnet 5: 55.5 (#26), GPT-6 Sol: 60.1 (#11)
| Benchmark | Claude Sonnet 5 | GPT-6 Sol |
|---|---|---|
| DeepSWE | 53.8% | 68.8% |
| FrontierCode | 42.7% | 49.3% |
| LMArena WebDev | 1541 | 1688 |
| SciCode | 54.3% | 57.6% |
| LMArena Coding | 1483 | 1447 |
| ALE-Bench | 1,463 | 2,462 |
| CursorBench | 34.1% | — |
| GSO | 37.3% | — |
| WeirdML | 68.8% | — |
Agentic & Tool Use Claude Sonnet 5 leads
Claude Sonnet 5: 42.8 (#18), GPT-6 Sol: 37.2 (#36)
| Benchmark | Claude Sonnet 5 | GPT-6 Sol |
|---|---|---|
| APEX-Agents | 54.5% | 54.3% |
| Vending-Bench 2 | 6,378 | 14,428 |
| GBAEval | 65.3% | — |
| GDP.pdf | — | 26.4% |
| LMArena Search | 1194 | — |
Reasoning GPT-6 Sol leads
Claude Sonnet 5: 49.1 (#39), GPT-6 Sol: 74.0 (#9)
| Benchmark | Claude Sonnet 5 | GPT-6 Sol |
|---|---|---|
| NYT Connections (extended) | 75.1% | 90.1% |
| CritPt | 16.9% | 30.9% |
| LMArena Hard Prompts | 1461 | 1418 |
| Mystery Game Puzzles | 35% | 56% |
| DTBench | 92.5% | 97.3% |
| LMCA | 50% | 59.1% |
| Epoch Capabilities Index | 156.21 | 162.72 |
| ARC-AGI-2 | — | 89.6% |
| SimpleBench | 60.6% | — |
| ARC-AGI-1 | — | 95.5% |
| Chess Puzzles | 35% | — |
| EBR-Bench | — | 53.3% |
| Surface Evolver Bench | 60% | — |
| Bench to the Future 3 | 0.14 | — |
| ForecastBench | 61.1 | — |
Math GPT-6 Sol leads
Claude Sonnet 5: 66.2 (#27), GPT-6 Sol: 87.2 (#7)
| Benchmark | Claude Sonnet 5 | GPT-6 Sol |
|---|---|---|
| FrontierMath (Tiers 1-3) | 65.6% | 89.8% |
| FrontierMath Tier 4 | 29.3% | 90% |
| OTIS Mock AIME 2024-2025 | 94.7% | 100% |
| ProofBench | 77% | 83% |
| LMArena Math | 1467 | 1402 |
Knowledge GPT-6 Sol leads
Claude Sonnet 5: 55.6 (#47), GPT-6 Sol: 64.8 (#15)
| Benchmark | Claude Sonnet 5 | GPT-6 Sol |
|---|---|---|
| GPQA Diamond | 90.5% | 94.3% |
| SimpleQA Verified | 33.7% | 60.7% |
| LMArena Expert | 1490 | 1439 |
| Vectara Hallucination Rate | — | 6.5% |
Multimodal GPT-6 Sol leads
Claude Sonnet 5: 42.4 (#31), GPT-6 Sol: 47.6 (#10)
| Benchmark | Claude Sonnet 5 | GPT-6 Sol |
|---|---|---|
| LMArena Vision | 1274 | 1245 |
| Blueprint-Bench 2 | 24.9% | 36.9% |
| Furniture Assembly | — | 58.3% |
| LMArena Document | 1466 | — |
Multilingual Claude Sonnet 5 leads
Claude Sonnet 5: 53.8 (#55), GPT-6 Sol: 50.5 (#118)
| Benchmark | Claude Sonnet 5 | GPT-6 Sol |
|---|---|---|
| LMArena Non-English | 1431 | 1385 |
| LMArena Chinese | 1477 | 1405 |
| LMArena French | 1460 | 1410 |
| LMArena German | 1440 | 1390 |
| LMArena Japanese | 1422 | 1385 |
| LMArena Korean | 1411 | 1341 |
| LMArena Russian | 1451 | 1401 |
| LMArena Spanish | 1437 | 1384 |
Instruction Following Claude Sonnet 5 leads
Claude Sonnet 5: 76.3 (#41), GPT-6 Sol: 74.5 (#94)
| Benchmark | Claude Sonnet 5 | GPT-6 Sol |
|---|---|---|
| LMArena Instruction Following | 1452 | 1412 |
Long Context Claude Sonnet 5 leads
Claude Sonnet 5: 44.8 (#55), GPT-6 Sol: 43.1 (#108)
| Benchmark | Claude Sonnet 5 | GPT-6 Sol |
|---|---|---|
| LMArena Longer Query | 1463 | 1411 |
Writing & Preference GPT-6 Sol leads
Claude Sonnet 5: 69.2 (#25), GPT-6 Sol: 71.9 (#18)
| Benchmark | Claude Sonnet 5 | GPT-6 Sol |
|---|---|---|
| LMArena Text | 1442 | 1395 |
| LMArena Creative Writing | 1416 | 1378 |
| EQ-Bench Creative Writing | 1794 | 2125 |
| LMArena Multi-Turn | 1454 | 1412 |
| EQ-Bench 4 | 1236 | — |
Frequently asked questions
Is Claude Sonnet 5 better than GPT-6 Sol?
GPT-6 Sol is the stronger model overall, scoring 61.8 to 54.6 on the Noometry Index.
Which is cheaper, Claude Sonnet 5 or GPT-6 Sol?
GPT-6 Sol is cheaper. It lists at $2 per million input tokens and $10 per million output tokens; Claude Sonnet 5 lists at $2 and $10.
Is Claude Sonnet 5 or GPT-6 Sol better for coding?
GPT-6 Sol scores higher on coding benchmarks: 60.1 versus 55.5 in the Noometry coding category.
Which has the bigger context window?
GPT-6 Sol does, with 1.05M tokens against 1M.
How many benchmarks do Claude Sonnet 5 and GPT-6 Sol share?
39 benchmarks have published results for both models. Claude Sonnet 5 has 51 scored results on Noometry and GPT-6 Sol has 45.