Model comparison
Claude Sonnet 5 vs GPT-5
Claude Sonnet 5 is the stronger model overall, scoring 54.6 to 50.9 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 7 categories and GPT-5 in 3 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in long context, where GPT-5 leads 69.5 to 44.8.
- The biggest single-benchmark swing is ProofBench: 77% for Claude Sonnet 5 and 18% for GPT-5.
- GPT-5 is cheaper at $1.25 / $10 per million input/output tokens, against $2 / $10 for Claude Sonnet 5.
- Claude Sonnet 5 accepts more context: 1M tokens versus 400K.
Side by side
| Claude Sonnet 5 | GPT-5 | |
|---|---|---|
| Provider | Anthropic | OpenAI |
| Noometry Index | 54.6 | 50.9 |
| Released | 2026-06-29 | 2025-08-07 |
| Weights | Proprietary | Proprietary |
| Context window | 1M | 400K |
| Max output | 128K | 128K |
| Input $ / M tokens | $2 | $1.25 |
| Output $ / M tokens | $10 | $10 |
| Results tracked | 51 | 69 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Claude Sonnet 5 leads
Claude Sonnet 5: 55.5 (#26), GPT-5: 50.3 (#47)
| Benchmark | Claude Sonnet 5 | GPT-5 |
|---|---|---|
| LMArena WebDev | 1541 | 1418 |
| SciCode | 54.3% | 42.9% |
| GSO | 37.3% | 6.9% |
| WeirdML | 68.8% | 60.7% |
| LMArena Coding | 1483 | 1436 |
| ALE-Bench | 1,463 | 1,162 |
| SWE-bench Verified | — | 73.6% |
| DeepSWE | 53.8% | — |
| FrontierCode | 42.7% | — |
| SWE-bench Verified (bash only) | — | 65% |
| Aider Polyglot | — | 88% |
| CursorBench | 34.1% | — |
| AlgoTune | — | 1.67 |
Agentic & Tool Use Claude Sonnet 5 leads
Claude Sonnet 5: 42.8 (#18), GPT-5: 33.1 (#56)
| Benchmark | Claude Sonnet 5 | GPT-5 |
|---|---|---|
| LMArena Search | 1194 | 1133 |
| Terminal-Bench | — | 49.6% |
| APEX-Agents | 54.5% | — |
| GDPval | — | 34.8% |
| Remote Labor Index | — | 1.7% |
| DeepResearch Bench | — | 49.6% |
| BALROG | — | 32.8% |
| GBAEval | 65.3% | — |
| METR Time Horizons | — | 69.6% |
| Vending-Bench 2 | 6,378 | — |
Reasoning Claude Sonnet 5 leads
Claude Sonnet 5: 49.1 (#39), GPT-5: 38.3 (#64)
| Benchmark | Claude Sonnet 5 | GPT-5 |
|---|---|---|
| SimpleBench | 60.6% | 56.7% |
| CritPt | 16.9% | 12.6% |
| Chess Puzzles | 35% | 37% |
| LMArena Hard Prompts | 1461 | 1416 |
| Mystery Game Puzzles | 35% | 23% |
| DTBench | 92.5% | 90.7% |
| LMCA | 50% | 40% |
| Epoch Capabilities Index | 156.21 | 150 |
| ForecastBench | 61.1 | 61.4 |
| ARC-AGI-2 | — | 9.9% |
| Kagi LLM Benchmark | — | 72.7% |
| NYT Connections (extended) | 75.1% | — |
| ARC-AGI-1 | — | 65.7% |
| EnigmaEval | — | 10.5% |
| EBR-Bench | — | 12.7% |
| Surface Evolver Bench | 60% | — |
| Bench to the Future 3 | 0.14 | — |
Math Claude Sonnet 5 leads
Claude Sonnet 5: 66.2 (#27), GPT-5: 55.0 (#44)
| Benchmark | Claude Sonnet 5 | GPT-5 |
|---|---|---|
| FrontierMath (Tiers 1-3) | 65.6% | 55.4% |
| FrontierMath Tier 4 | 29.3% | 22% |
| OTIS Mock AIME 2024-2025 | 94.7% | 91.4% |
| ProofBench | 77% | 18% |
| LMArena Math | 1467 | 1407 |
| Omni-MATH | — | 64.7% |
| MATH Level 5 | — | 98.1% |
| FrontierMath (Feb 2025 set) | — | 32.4% |
| FrontierMath Tier 4 (v1) | — | 12.5% |
Knowledge Too close to call
Claude Sonnet 5: 55.6 (#47), GPT-5: 56.6 (#43)
| Benchmark | Claude Sonnet 5 | GPT-5 |
|---|---|---|
| GPQA Diamond | 90.5% | 86.2% |
| SimpleQA Verified | 33.7% | 50.1% |
| LMArena Expert | 1490 | 1419 |
| Humanity's Last Exam | — | 25.3% |
| MMLU-Pro | — | 86.3% |
| Confabulations | — | 10.3% |
| Vectara Hallucination Rate | — | 14.7% |
| GPQA (HELM) | — | 79.2% |
Multimodal GPT-5 leads
Claude Sonnet 5: 42.4 (#31), GPT-5: 46.8 (#13)
| Benchmark | Claude Sonnet 5 | GPT-5 |
|---|---|---|
| LMArena Vision | 1274 | 1232 |
| GeoBench | — | 81% |
| VPCT | — | 66% |
| Blueprint-Bench 2 | 24.9% | — |
| LMArena Document | 1466 | — |
Multilingual Claude Sonnet 5 leads
Claude Sonnet 5: 53.8 (#55), GPT-5: 51.4 (#110)
| Benchmark | Claude Sonnet 5 | GPT-5 |
|---|---|---|
| LMArena Non-English | 1431 | 1397 |
| LMArena Chinese | 1477 | 1422 |
| LMArena French | 1460 | 1410 |
| LMArena German | 1440 | 1416 |
| LMArena Japanese | 1422 | 1409 |
| LMArena Korean | 1411 | 1360 |
| LMArena Russian | 1451 | 1406 |
| LMArena Spanish | 1437 | 1399 |
Instruction Following Claude Sonnet 5 leads
Claude Sonnet 5: 76.3 (#41), GPT-5: 73.8 (#113)
| Benchmark | Claude Sonnet 5 | GPT-5 |
|---|---|---|
| LMArena Instruction Following | 1452 | 1388 |
| IFEval | — | 87.5% |
Long Context GPT-5 leads
Claude Sonnet 5: 44.8 (#55), GPT-5: 69.5 (#2)
| Benchmark | Claude Sonnet 5 | GPT-5 |
|---|---|---|
| LMArena Longer Query | 1463 | 1399 |
| Fiction.LiveBench | — | 97.2% |
Writing & Preference Claude Sonnet 5 leads
Claude Sonnet 5: 69.2 (#25), GPT-5: 63.4 (#65)
| Benchmark | Claude Sonnet 5 | GPT-5 |
|---|---|---|
| LMArena Text | 1442 | 1406 |
| LMArena Creative Writing | 1416 | 1365 |
| EQ-Bench Creative Writing | 1794 | 1627 |
| LMArena Multi-Turn | 1454 | 1426 |
| Short-Story Creative Writing | — | 86% |
| WildBench | — | 85.7% |
| EQ-Bench 4 | 1236 | — |
Frequently asked questions
Is Claude Sonnet 5 better than GPT-5?
Claude Sonnet 5 is the stronger model overall, scoring 54.6 to 50.9 on the Noometry Index.
Which is cheaper, Claude Sonnet 5 or GPT-5?
GPT-5 is cheaper. It lists at $1.25 per million input tokens and $10 per million output tokens; Claude Sonnet 5 lists at $2 and $10.
Is Claude Sonnet 5 or GPT-5 better for coding?
Claude Sonnet 5 scores higher on coding benchmarks: 55.5 versus 50.3 in the Noometry coding category.
Which has the bigger context window?
Claude Sonnet 5 does, with 1M tokens against 400K.
How many benchmarks do Claude Sonnet 5 and GPT-5 share?
39 benchmarks have published results for both models. Claude Sonnet 5 has 51 scored results on Noometry and GPT-5 has 69.