Model comparison
Claude Sonnet 4.6 vs GPT-6 Sol
GPT-6 Sol is the stronger model overall, scoring 61.8 to 50.3 on the Noometry Index.
Last verified . 41 shared benchmarks.
Summary
- They share 41 benchmarks with published results for both. Claude Sonnet 4.6 scores higher in 4 categories and GPT-6 Sol in 6 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in math, where GPT-6 Sol leads 87.2 to 52.9.
- The biggest single-benchmark swing is Mystery Game Puzzles: 16% for Claude Sonnet 4.6 and 56% for GPT-6 Sol.
- GPT-6 Sol is cheaper at $2 / $10 per million input/output tokens, against $3 / $15 for Claude Sonnet 4.6.
- GPT-6 Sol accepts more context: 1.05M tokens versus 1M.
Side by side
| Claude Sonnet 4.6 | GPT-6 Sol | |
|---|---|---|
| Provider | Anthropic | OpenAI |
| Noometry Index | 50.3 | 61.8 |
| Released | 2026-02-17 | 2026-09-22 |
| Weights | Proprietary | Proprietary |
| Context window | 1M | 1.05M |
| Max output | 128K | 128K |
| Input $ / M tokens | $3 | $2 |
| Output $ / M tokens | $15 | $10 |
| Results tracked | 57 | 45 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-6 Sol leads
Claude Sonnet 4.6: 46.3 (#67), GPT-6 Sol: 60.1 (#11)
| Benchmark | Claude Sonnet 4.6 | GPT-6 Sol |
|---|---|---|
| DeepSWE | 29.9% | 68.8% |
| FrontierCode | 24.3% | 49.3% |
| LMArena WebDev | 1522 | 1688 |
| SciCode | 46.8% | 57.6% |
| LMArena Coding | 1504 | 1447 |
| ALE-Bench | 1,327 | 2,462 |
| SWE-bench Verified | 75.2% | — |
| WeirdML | 66.1% | — |
Agentic & Tool Use Claude Sonnet 4.6 leads
Claude Sonnet 4.6: 39.1 (#28), GPT-6 Sol: 37.2 (#36)
| Benchmark | Claude Sonnet 4.6 | GPT-6 Sol |
|---|---|---|
| APEX-Agents | 43% | 54.3% |
| GDP.pdf | 18% | 26.4% |
| Vending-Bench 2 | 7,204 | 14,428 |
| Terminal-Bench | 53.4% | — |
| OSWorld 2.0 | 9.3% | — |
| DeepResearch Bench | 54.9% | — |
| OSWorld | 72.1% | — |
| ExploitBench | 23.6% | — |
| GBAEval | 48.8% | — |
| LMArena Search | 1221 | — |
Reasoning GPT-6 Sol leads
Claude Sonnet 4.6: 46.1 (#45), GPT-6 Sol: 74.0 (#9)
| Benchmark | Claude Sonnet 4.6 | GPT-6 Sol |
|---|---|---|
| ARC-AGI-2 | 60.4% | 89.6% |
| NYT Connections (extended) | 80.9% | 90.1% |
| ARC-AGI-1 | 86.5% | 95.5% |
| CritPt | 3.1% | 30.9% |
| LMArena Hard Prompts | 1484 | 1418 |
| Mystery Game Puzzles | 16% | 56% |
| DTBench | 89.9% | 97.3% |
| LMCA | 46.5% | 59.1% |
| Epoch Capabilities Index | 152.24 | 162.72 |
| Chess Puzzles | 13% | — |
| Thematic Generalization | 76.3% | — |
| EBR-Bench | — | 53.3% |
| ForecastBench | 62 | — |
Math GPT-6 Sol leads
Claude Sonnet 4.6: 52.9 (#49), GPT-6 Sol: 87.2 (#7)
| Benchmark | Claude Sonnet 4.6 | GPT-6 Sol |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 85.8% | 100% |
| ProofBench | 45% | 83% |
| LMArena Math | 1462 | 1402 |
| FrontierMath (Tiers 1-3) | — | 89.8% |
| FrontierMath Tier 4 | — | 90% |
| FrontierMath (Feb 2025 set) | 32.4% | — |
| FrontierMath Tier 4 (v1) | 8.3% | — |
Knowledge GPT-6 Sol leads
Claude Sonnet 4.6: 51.7 (#65), GPT-6 Sol: 64.8 (#15)
| Benchmark | Claude Sonnet 4.6 | GPT-6 Sol |
|---|---|---|
| GPQA Diamond | 87.4% | 94.3% |
| SimpleQA Verified | 35.5% | 60.7% |
| Vectara Hallucination Rate | 10.6% | 6.5% |
| LMArena Expert | 1500 | 1439 |
Multimodal GPT-6 Sol leads
Claude Sonnet 4.6: 38.0 (#68), GPT-6 Sol: 47.6 (#10)
| Benchmark | Claude Sonnet 4.6 | GPT-6 Sol |
|---|---|---|
| LMArena Vision | 1283 | 1245 |
| Blueprint-Bench 2 | 6.7% | 36.9% |
| Furniture Assembly | — | 58.3% |
| LMArena Document | 1482 | — |
Multilingual Claude Sonnet 4.6 leads
Claude Sonnet 4.6: 54.4 (#41), GPT-6 Sol: 50.5 (#118)
| Benchmark | Claude Sonnet 4.6 | GPT-6 Sol |
|---|---|---|
| LMArena Non-English | 1440 | 1385 |
| LMArena Chinese | 1491 | 1405 |
| LMArena French | 1465 | 1410 |
| LMArena German | 1428 | 1390 |
| LMArena Japanese | 1420 | 1385 |
| LMArena Korean | 1411 | 1341 |
| LMArena Russian | 1440 | 1401 |
| LMArena Spanish | 1464 | 1384 |
Instruction Following Claude Sonnet 4.6 leads
Claude Sonnet 4.6: 77.4 (#25), GPT-6 Sol: 74.5 (#94)
| Benchmark | Claude Sonnet 4.6 | GPT-6 Sol |
|---|---|---|
| LMArena Instruction Following | 1475 | 1412 |
Long Context Claude Sonnet 4.6 leads
Claude Sonnet 4.6: 45.3 (#44), GPT-6 Sol: 43.1 (#108)
| Benchmark | Claude Sonnet 4.6 | GPT-6 Sol |
|---|---|---|
| LMArena Longer Query | 1479 | 1411 |
Writing & Preference GPT-6 Sol leads
Claude Sonnet 4.6: 70.2 (#22), GPT-6 Sol: 71.9 (#18)
| Benchmark | Claude Sonnet 4.6 | GPT-6 Sol |
|---|---|---|
| LMArena Text | 1458 | 1395 |
| LMArena Creative Writing | 1435 | 1378 |
| EQ-Bench Creative Writing | 1810 | 2125 |
| LMArena Multi-Turn | 1464 | 1412 |
| EQ-Bench 4 | 1207 | — |
Frequently asked questions
Is Claude Sonnet 4.6 better than GPT-6 Sol?
GPT-6 Sol is the stronger model overall, scoring 61.8 to 50.3 on the Noometry Index.
Which is cheaper, Claude Sonnet 4.6 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 4.6 lists at $3 and $15.
Is Claude Sonnet 4.6 or GPT-6 Sol better for coding?
GPT-6 Sol scores higher on coding benchmarks: 60.1 versus 46.3 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 4.6 and GPT-6 Sol share?
41 benchmarks have published results for both models. Claude Sonnet 4.6 has 57 scored results on Noometry and GPT-6 Sol has 45.