Model comparison
Claude 3.7 Sonnet vs GPT-5.4 nano
GPT-5.4 nano is the stronger model overall, scoring 41.9 to 39.5 on the Noometry Index.
Last verified . 25 shared benchmarks.
Summary
- They share 25 benchmarks with published results for both. Claude 3.7 Sonnet scores higher in 2 categories and GPT-5.4 nano in 7 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in long context, where Claude 3.7 Sonnet leads 50.3 to 41.6.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 57.8% for Claude 3.7 Sonnet and 87.8% for GPT-5.4 nano.
Side by side
| Claude 3.7 Sonnet | GPT-5.4 nano | |
|---|---|---|
| Provider | Anthropic | OpenAI |
| Noometry Index | 39.5 | 41.9 |
| Released | 2025-02-24 | 2026-03-17 |
| Weights | Proprietary | Proprietary |
| Context window | — | 400K |
| Max output | — | 128K |
| Input $ / M tokens | — | $0.20 |
| Output $ / M tokens | — | $1.25 |
| Results tracked | 58 | 40 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-5.4 nano leads
Claude 3.7 Sonnet: 40.6 (#136), GPT-5.4 nano: 43.6 (#84)
| Benchmark | Claude 3.7 Sonnet | GPT-5.4 nano |
|---|---|---|
| LMArena Coding | 1361 | 1405 |
| SWE-bench Verified | 61% | — |
| SWE-bench Verified (bash only) | 52.8% | — |
| Aider Polyglot | 64.9% | — |
| SciCode | — | 46.9% |
| GSO | 3.8% | — |
| WeirdML | — | 49.2% |
| LiveBench Coding | 74.5% | — |
| CadEval | 54% | — |
| ALE-Bench | — | 1,005 |
Agentic & Tool Use Not comparable
Claude 3.7 Sonnet: 34.1 (#50), GPT-5.4 nano: —
| Benchmark | Claude 3.7 Sonnet | GPT-5.4 nano |
|---|---|---|
| TheAgentCompany | 30.9% | — |
| Cybench | 20% | — |
| DeepResearch Bench | 43.6% | — |
| OSWorld | 35.8% | — |
| METR Time Horizons | 60% | — |
Reasoning GPT-5.4 nano leads
Claude 3.7 Sonnet: 18.6 (#277), GPT-5.4 nano: 23.7 (#173)
| Benchmark | Claude 3.7 Sonnet | GPT-5.4 nano |
|---|---|---|
| ARC-AGI-2 | 0.9% | 5.7% |
| ARC-AGI-1 | 28.6% | 51.5% |
| LMArena Hard Prompts | 1333 | 1381 |
| Epoch Capabilities Index | 141.16 | 145.81 |
| ForecastBench | 61.8 | 57.3 |
| SimpleBench | 46.4% | — |
| Kagi LLM Benchmark | — | 39.7% |
| CritPt | — | 9.3% |
| Chess Puzzles | — | 30% |
| EnigmaEval | 4.2% | — |
| LiveBench Reasoning | 87.8% | — |
| Mystery Game Puzzles | — | 9% |
| DTBench | — | 80.3% |
| LiveBench Data Analysis | 74% | — |
| LMCA | — | 36.9% |
| LiveBench | 76.1% | — |
Math GPT-5.4 nano leads
Claude 3.7 Sonnet: 37.5 (#153), GPT-5.4 nano: 40.9 (#88)
| Benchmark | Claude 3.7 Sonnet | GPT-5.4 nano |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 57.8% | 87.8% |
| LMArena Math | 1337 | 1406 |
| FrontierMath (Feb 2025 set) | 4.1% | 25.9% |
| FrontierMath (Tiers 1-3) | — | 44.9% |
| FrontierMath Tier 4 | — | 12.2% |
| ProofBench | — | 5% |
| Omni-MATH | 33% | — |
| LiveBench Math | 79% | — |
| MATH Level 5 | 91.2% | — |
| FrontierMath Tier 4 (v1) | — | 6.3% |
Knowledge GPT-5.4 nano leads
Claude 3.7 Sonnet: 39.8 (#130), GPT-5.4 nano: 41.9 (#103)
| Benchmark | Claude 3.7 Sonnet | GPT-5.4 nano |
|---|---|---|
| GPQA Diamond | 79.7% | 78.5% |
| LMArena Expert | 1321 | 1396 |
| Humanity's Last Exam | 8% | — |
| SimpleQA Verified | — | 11.7% |
| MMLU-Pro | 78.4% | — |
| Confabulations | 14.7% | — |
| Vectara Hallucination Rate | — | 3.1% |
| GPQA (HELM) | 60.8% | — |
Multimodal GPT-5.4 nano leads
Claude 3.7 Sonnet: 33.7 (#95), GPT-5.4 nano: 36.7 (#78)
| Benchmark | Claude 3.7 Sonnet | GPT-5.4 nano |
|---|---|---|
| LMArena Vision | 1169 | 1196 |
| GeoBench | 68% | — |
| VPCT | 39% | — |
| SpatialViz-Bench | 33.9% | — |
Multilingual GPT-5.4 nano leads
Claude 3.7 Sonnet: 44.1 (#179), GPT-5.4 nano: 48.6 (#140)
| Benchmark | Claude 3.7 Sonnet | GPT-5.4 nano |
|---|---|---|
| LMArena Non-English | 1296 | 1359 |
| LMArena Chinese | 1299 | 1392 |
| LMArena French | 1303 | 1396 |
| LMArena German | 1301 | 1367 |
| LMArena Japanese | 1267 | 1343 |
| LMArena Korean | 1249 | 1320 |
| LMArena Russian | 1311 | 1363 |
| LMArena Spanish | 1298 | 1371 |
Instruction Following Too close to call
Claude 3.7 Sonnet: 72.9 (#125), GPT-5.4 nano: 71.9 (#144)
| Benchmark | Claude 3.7 Sonnet | GPT-5.4 nano |
|---|---|---|
| LMArena Instruction Following | 1352 | 1362 |
| LiveBench Instruction Following | 81.3% | — |
| IFEval | 83.4% | — |
Long Context Claude 3.7 Sonnet leads
Claude 3.7 Sonnet: 50.3 (#10), GPT-5.4 nano: 41.6 (#137)
| Benchmark | Claude 3.7 Sonnet | GPT-5.4 nano |
|---|---|---|
| LMArena Longer Query | 1373 | 1366 |
| Fiction.LiveBench | 83.3% | — |
Writing & Preference GPT-5.4 nano leads
Claude 3.7 Sonnet: 54.4 (#150), GPT-5.4 nano: 55.7 (#142)
| Benchmark | Claude 3.7 Sonnet | GPT-5.4 nano |
|---|---|---|
| LMArena Text | 1314 | 1372 |
| LMArena Creative Writing | 1332 | 1314 |
| LMArena Multi-Turn | 1339 | 1382 |
| Short-Story Creative Writing | 81.1% | — |
| EQ-Bench Creative Writing | 1412 | — |
| WildBench | 81.4% | — |
| LiveBench Language | 59.9% | — |
Frequently asked questions
Is Claude 3.7 Sonnet better than GPT-5.4 nano?
GPT-5.4 nano is the stronger model overall, scoring 41.9 to 39.5 on the Noometry Index.
Is Claude 3.7 Sonnet or GPT-5.4 nano better for coding?
GPT-5.4 nano scores higher on coding benchmarks: 43.6 versus 40.6 in the Noometry coding category.
How many benchmarks do Claude 3.7 Sonnet and GPT-5.4 nano share?
25 benchmarks have published results for both models. Claude 3.7 Sonnet has 58 scored results on Noometry and GPT-5.4 nano has 40.