Model comparison
Claude Sonnet 4.6 vs GPT-4.5
Claude Sonnet 4.6 is the stronger model overall, scoring 50.1 to 36.9 on the Noometry Index.
Last verified . 25 shared benchmarks.
Summary
- They share 25 benchmarks with published results for both. Claude Sonnet 4.6 scores higher in 10 categories and GPT-4.5 in 0 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Claude Sonnet 4.6 leads 46.1 to 14.1.
- The biggest single-benchmark swing is ARC-AGI-1: 86.5% for Claude Sonnet 4.6 and 10.3% for GPT-4.5.
Side by side
| Claude Sonnet 4.6 | GPT-4.5 | |
|---|---|---|
| Provider | Anthropic | OpenAI |
| Noometry Index | 50.1 | 36.9 |
| Released | 2026-02-17 | 2025-02-27 |
| Weights | Proprietary | Proprietary |
| Context window | 1M | — |
| Max output | 128K | — |
| Input $ / M tokens | $3 | — |
| Output $ / M tokens | $15 | — |
| Results tracked | 57 | 42 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Claude Sonnet 4.6 leads
Claude Sonnet 4.6: 46.0 (#62), GPT-4.5: 41.8 (#102)
| Benchmark | Claude Sonnet 4.6 | GPT-4.5 |
|---|---|---|
| WeirdML | 66.1% | 39.4% |
| LMArena Coding | 1504 | 1397 |
| SWE-bench Verified | 75.2% | — |
| DeepSWE | 29.9% | — |
| FrontierCode | 24.3% | — |
| Aider Polyglot | — | 44.9% |
| LMArena WebDev | 1521 | — |
| SciCode | 46.8% | — |
| LiveBench Coding | — | 75.2% |
| ALE-Bench | 1,327 | — |
Agentic & Tool Use Claude Sonnet 4.6 leads
Claude Sonnet 4.6: 39.1 (#26), GPT-4.5: 27.6 (#92)
| Benchmark | Claude Sonnet 4.6 | GPT-4.5 |
|---|---|---|
| Terminal-Bench | 53.4% | — |
| APEX-Agents | 43% | — |
| OSWorld 2.0 | 9.3% | — |
| Cybench | — | 17.5% |
| DeepResearch Bench | 54.9% | — |
| OSWorld | 72.1% | — |
| ExploitBench | 23.6% | — |
| GBAEval | 48.8% | — |
| GDP.pdf | 18% | — |
| LMArena Search | 1221 | — |
| Vending-Bench 2 | 7,204 | — |
Reasoning Claude Sonnet 4.6 leads
Claude Sonnet 4.6: 46.1 (#40), GPT-4.5: 14.1 (#320)
| Benchmark | Claude Sonnet 4.6 | GPT-4.5 |
|---|---|---|
| ARC-AGI-2 | 60.4% | 0.8% |
| ARC-AGI-1 | 86.5% | 10.3% |
| LMArena Hard Prompts | 1485 | 1404 |
| Epoch Capabilities Index | 152.25 | 136.74 |
| ForecastBench | 62 | 61.7 |
| SimpleBench | — | 34.5% |
| NYT Connections (extended) | 80.9% | — |
| CritPt | 3.1% | — |
| Chess Puzzles | 13% | — |
| EnigmaEval | — | 3.2% |
| Thematic Generalization | 76.3% | — |
| LiveBench Reasoning | — | 71.1% |
| Mystery Game Puzzles | 16% | — |
| DTBench | 89.9% | — |
| LiveBench Data Analysis | — | 64.3% |
| LMCA | 46.5% | — |
| LiveBench | — | 69% |
Math Claude Sonnet 4.6 leads
Claude Sonnet 4.6: 51.9 (#41), GPT-4.5: 31.0 (#200)
| Benchmark | Claude Sonnet 4.6 | GPT-4.5 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 85.8% | 37.8% |
| LMArena Math | 1463 | 1412 |
| ProofBench | 45% | — |
| LiveBench Math | — | 69.3% |
| MATH Level 5 | — | 78.6% |
| FrontierMath (Feb 2025 set) | 32.4% | — |
| FrontierMath Tier 4 (v1) | 8.3% | — |
Knowledge Claude Sonnet 4.6 leads
Claude Sonnet 4.6: 51.6 (#60), GPT-4.5: 32.4 (#205)
| Benchmark | Claude Sonnet 4.6 | GPT-4.5 |
|---|---|---|
| GPQA Diamond | 87.4% | 68.7% |
| LMArena Expert | 1500 | 1394 |
| Humanity's Last Exam | — | 5.4% |
| SimpleQA Verified | 35.5% | — |
| Confabulations | — | 13.6% |
| Vectara Hallucination Rate | 10.6% | — |
Multimodal Claude Sonnet 4.6 leads
Claude Sonnet 4.6: 38.0 (#61), GPT-4.5: 37.0 (#68)
| Benchmark | Claude Sonnet 4.6 | GPT-4.5 |
|---|---|---|
| LMArena Vision | 1283 | 1195 |
| VPCT | — | 45% |
| Blueprint-Bench 2 | 6.7% | — |
| LMArena Document | 1482 | — |
Multilingual Claude Sonnet 4.6 leads
Claude Sonnet 4.6: 54.6 (#39), GPT-4.5: 52.7 (#78)
| Benchmark | Claude Sonnet 4.6 | GPT-4.5 |
|---|---|---|
| LMArena Non-English | 1440 | 1413 |
| LMArena Chinese | 1491 | 1421 |
| LMArena French | 1465 | 1417 |
| LMArena German | 1428 | 1457 |
| LMArena Japanese | 1419 | 1416 |
| LMArena Korean | 1411 | 1392 |
| LMArena Russian | 1440 | 1419 |
| LMArena Spanish | 1465 | — |
Instruction Following Claude Sonnet 4.6 leads
Claude Sonnet 4.6: 77.6 (#23), GPT-4.5: 72.8 (#129)
| Benchmark | Claude Sonnet 4.6 | GPT-4.5 |
|---|---|---|
| LMArena Instruction Following | 1476 | 1404 |
| LiveBench Instruction Following | — | 72.3% |
Long Context Claude Sonnet 4.6 leads
Claude Sonnet 4.6: 45.3 (#40), GPT-4.5: 40.4 (#150)
| Benchmark | Claude Sonnet 4.6 | GPT-4.5 |
|---|---|---|
| LMArena Longer Query | 1479 | 1406 |
| Fiction.LiveBench | — | 63.9% |
Writing & Preference Claude Sonnet 4.6 leads
Claude Sonnet 4.6: 70.4 (#22), GPT-4.5: 57.1 (#129)
| Benchmark | Claude Sonnet 4.6 | GPT-4.5 |
|---|---|---|
| LMArena Text | 1458 | 1417 |
| LMArena Creative Writing | 1435 | 1394 |
| EQ-Bench Creative Writing | 1810 | 1258 |
| LMArena Multi-Turn | 1464 | 1444 |
| Short-Story Creative Writing | — | 75.6% |
| EQ-Bench 4 | 1207 | — |
| LiveBench Language | — | 61.5% |
Frequently asked questions
Is Claude Sonnet 4.6 better than GPT-4.5?
Claude Sonnet 4.6 is the stronger model overall, scoring 50.1 to 36.9 on the Noometry Index.
Is Claude Sonnet 4.6 or GPT-4.5 better for coding?
Claude Sonnet 4.6 scores higher on coding benchmarks: 46.0 versus 41.8 in the Noometry coding category.
How many benchmarks do Claude Sonnet 4.6 and GPT-4.5 share?
25 benchmarks have published results for both models. Claude Sonnet 4.6 has 57 scored results on Noometry and GPT-4.5 has 42.