Model comparison
Claude Sonnet 4 vs GPT-4o
Claude Sonnet 4 is the stronger model overall, scoring 40.6 to 28.5 on the Noometry Index.
Last verified . 50 shared benchmarks.
Summary
- They share 50 benchmarks with published results for both. Claude Sonnet 4 scores higher in 8 categories and GPT-4o in 2 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in math, where Claude Sonnet 4 leads 41.5 to 10.2.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 71.1% for Claude Sonnet 4 and 6.4% for GPT-4o.
- GPT-4o is cheaper at $2.50 / $10 per million input/output tokens, against $3 / $15 for Claude Sonnet 4.
- Claude Sonnet 4 accepts more context: 200K tokens versus 128K.
Side by side
| Claude Sonnet 4 | GPT-4o | |
|---|---|---|
| Provider | Anthropic | OpenAI |
| Noometry Index | 40.6 | 28.5 |
| Released | 2025-05-22 | 2024-05-13 |
| Weights | Proprietary | Proprietary |
| Context window | 200K | 128K |
| Max output | 64K | 16K |
| Input $ / M tokens | $3 | $2.50 |
| Output $ / M tokens | $15 | $10 |
| Results tracked | 58 | 72 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Claude Sonnet 4 leads
Claude Sonnet 4: 44.0 (#70), GPT-4o: 25.1 (#323)
| Benchmark | Claude Sonnet 4 | GPT-4o |
|---|---|---|
| SWE-bench Verified (bash only) | 64.9% | 21.6% |
| Aider Polyglot | 61.3% | 45.3% |
| GSO | 4.9% | 0% |
| WeirdML | 46.1% | 25.1% |
| LMArena Coding | 1414 | 1297 |
| SWE-bench Verified | — | 31% |
| SciCode | 40% | — |
| BigCodeBench Instruct | — | 51.1% |
| LiveBench Coding | — | 51.4% |
| BigCodeBench Complete | — | 61.1% |
| CadEval | — | 26% |
| ALE-Bench | 655.35 | — |
| HumanEval+ | — | 87.2% |
| MBPP+ | — | 72.2% |
Agentic & Tool Use Claude Sonnet 4 leads
Claude Sonnet 4: 38.3 (#29), GPT-4o: 20.8 (#133)
| Benchmark | Claude Sonnet 4 | GPT-4o |
|---|---|---|
| TheAgentCompany | 33.1% | 8.6% |
| Cybench | 35% | 12.5% |
| METR Time Horizons | 62% | 40.8% |
| GDPval | — | 9.9% |
| DeepResearch Bench | 46.6% | — |
| OSWorld | 43.9% | — |
| BALROG | — | 32.3% |
| LMArena Search | — | 1006 |
Reasoning Claude Sonnet 4 leads
Claude Sonnet 4: 23.0 (#169), GPT-4o: 9.4 (#338)
| Benchmark | Claude Sonnet 4 | GPT-4o |
|---|---|---|
| ARC-AGI-2 | 5.9% | 0% |
| SimpleBench | 45.5% | 17.8% |
| ARC-AGI-1 | 40% | 4.5% |
| CritPt | 0.3% | 0% |
| EnigmaEval | 3.1% | 0.8% |
| LMArena Hard Prompts | 1372 | 1281 |
| DTBench | 77.1% | 64.5% |
| LMCA | 29% | 16.6% |
| Epoch Capabilities Index | 141.69 | 128.97 |
| ForecastBench | 60.2 | 57.7 |
| Kagi LLM Benchmark | 73% | — |
| Chess Puzzles | — | 13% |
| LiveBench Reasoning | — | 55.8% |
| LiveBench Data Analysis | — | 60.9% |
| LiveBench | — | 55.3% |
Math Claude Sonnet 4 leads
Claude Sonnet 4: 41.5 (#71), GPT-4o: 10.2 (#303)
| Benchmark | Claude Sonnet 4 | GPT-4o |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 71.1% | 6.4% |
| Omni-MATH | 60.2% | 29.3% |
| LMArena Math | 1375 | 1285 |
| MATH Level 5 | 84.4% | 53.3% |
| FrontierMath (Feb 2025 set) | 4.1% | 0.3% |
| FrontierMath (Tiers 1-3) | — | 0.4% |
| LiveBench Math | — | 49.5% |
| FrontierMath Tier 4 (v1) | 0% | — |
Knowledge Claude Sonnet 4 leads
Claude Sonnet 4: 41.6 (#99), GPT-4o: 28.8 (#235)
| Benchmark | Claude Sonnet 4 | GPT-4o |
|---|---|---|
| GPQA Diamond | 79.2% | 49.2% |
| Humanity's Last Exam | 7.8% | 2.7% |
| MMLU-Pro | 84.3% | 71.3% |
| Confabulations | 13.2% | 15.3% |
| Vectara Hallucination Rate | 10.3% | 9.6% |
| GPQA (HELM) | 70.6% | 52% |
| LMArena Expert | 1371 | 1250 |
| SimpleQA Verified | — | 26% |
| MMLU | — | 88.1% |
Multimodal GPT-4o leads
Claude Sonnet 4: 25.6 (#116), GPT-4o: 33.9 (#85)
| Benchmark | Claude Sonnet 4 | GPT-4o |
|---|---|---|
| LMArena Vision | 1191 | 1137 |
| GeoBench | 37% | 71% |
| VPCT | 34% | 40% |
| Video-MME | — | 71.9% |
| MindCube | 44.8% | — |
| ScienceQA | — | 88.5% |
Multilingual Claude Sonnet 4 leads
Claude Sonnet 4: 46.9 (#151), GPT-4o: 43.4 (#181)
| Benchmark | Claude Sonnet 4 | GPT-4o |
|---|---|---|
| LMArena Non-English | 1333 | 1283 |
| LMArena Chinese | 1351 | 1277 |
| LMArena French | 1364 | 1303 |
| LMArena German | 1331 | 1282 |
| LMArena Japanese | 1301 | 1257 |
| LMArena Korean | 1293 | 1234 |
| LMArena Russian | 1355 | 1286 |
| LMArena Spanish | 1356 | 1292 |
Instruction Following Claude Sonnet 4 leads
Claude Sonnet 4: 71.8 (#141), GPT-4o: 66.8 (#202)
| Benchmark | Claude Sonnet 4 | GPT-4o |
|---|---|---|
| IFEval | 84% | 81.7% |
| LMArena Instruction Following | 1375 | 1278 |
| LiveBench Instruction Following | — | 68.6% |
Long Context GPT-4o leads
Claude Sonnet 4: 33.6 (#254), GPT-4o: 39.4 (#174)
| Benchmark | Claude Sonnet 4 | GPT-4o |
|---|---|---|
| Fiction.LiveBench | 46.9% | 66.7% |
| LMArena Longer Query | 1398 | 1289 |
Writing & Preference Claude Sonnet 4 leads
Claude Sonnet 4: 57.3 (#127), GPT-4o: 52.8 (#160)
| Benchmark | Claude Sonnet 4 | GPT-4o |
|---|---|---|
| LMArena Text | 1351 | 1300 |
| LMArena Creative Writing | 1345 | 1291 |
| Short-Story Creative Writing | 81.4% | 81.8% |
| WildBench | 83.8% | 82.8% |
| LMArena Multi-Turn | 1376 | 1302 |
| EQ-Bench Creative Writing | 1483 | — |
| LiveBench Language | — | 47.6% |
Frequently asked questions
Is Claude Sonnet 4 better than GPT-4o?
Claude Sonnet 4 is the stronger model overall, scoring 40.6 to 28.5 on the Noometry Index.
Which is cheaper, Claude Sonnet 4 or GPT-4o?
GPT-4o is cheaper. It lists at $2.50 per million input tokens and $10 per million output tokens; Claude Sonnet 4 lists at $3 and $15.
Is Claude Sonnet 4 or GPT-4o better for coding?
Claude Sonnet 4 scores higher on coding benchmarks: 44.0 versus 25.1 in the Noometry coding category.
Which has the bigger context window?
Claude Sonnet 4 does, with 200K tokens against 128K.
How many benchmarks do Claude Sonnet 4 and GPT-4o share?
50 benchmarks have published results for both models. Claude Sonnet 4 has 58 scored results on Noometry and GPT-4o has 72.