Model comparison
Claude 3.5 Sonnet vs GPT-4o
Claude 3.5 Sonnet is the stronger model overall, scoring 34.4 to 28.5 on the Noometry Index.
Last verified . 58 shared benchmarks.
Summary
- They share 58 benchmarks with published results for both. Claude 3.5 Sonnet scores higher in 8 categories and GPT-4o in 2 categories; 6 gaps are clear of the uncertainty.
- The widest gap is in coding, where Claude 3.5 Sonnet leads 39.5 to 25.1.
- The biggest single-benchmark swing is SimpleBench: 41.4% for Claude 3.5 Sonnet and 17.8% for GPT-4o.
Side by side
| Claude 3.5 Sonnet | GPT-4o | |
|---|---|---|
| Provider | Anthropic | OpenAI |
| Noometry Index | 34.4 | 28.5 |
| Released | 2024-06-20 | 2024-05-13 |
| Weights | Proprietary | Proprietary |
| Context window | — | 128K |
| Max output | — | 16K |
| Input $ / M tokens | — | $2.50 |
| Output $ / M tokens | — | $10 |
| Results tracked | 60 | 72 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Claude 3.5 Sonnet leads
Claude 3.5 Sonnet: 39.5 (#145), GPT-4o: 25.1 (#323)
| Benchmark | Claude 3.5 Sonnet | GPT-4o |
|---|---|---|
| Aider Polyglot | 51.6% | 45.3% |
| GSO | 4.6% | 0% |
| WeirdML | 40% | 25.1% |
| BigCodeBench Instruct | 46.8% | 51.1% |
| LiveBench Coding | 67.1% | 51.4% |
| LMArena Coding | 1343 | 1297 |
| BigCodeBench Complete | 58.6% | 61.1% |
| CadEval | 48% | 26% |
| HumanEval+ | 81.7% | 87.2% |
| MBPP+ | 74.3% | 72.2% |
| SWE-bench Verified | — | 31% |
| SWE-bench Verified (bash only) | — | 21.6% |
Agentic & Tool Use Claude 3.5 Sonnet leads
Claude 3.5 Sonnet: 32.1 (#61), GPT-4o: 20.8 (#133)
| Benchmark | Claude 3.5 Sonnet | GPT-4o |
|---|---|---|
| TheAgentCompany | 24% | 8.6% |
| Cybench | 17.5% | 12.5% |
| BALROG | 32.6% | 32.3% |
| METR Time Horizons | 45.2% | 40.8% |
| GDPval | — | 9.9% |
| LMArena Search | — | 1006 |
Reasoning Claude 3.5 Sonnet leads
Claude 3.5 Sonnet: 22.5 (#176), GPT-4o: 9.4 (#338)
| Benchmark | Claude 3.5 Sonnet | GPT-4o |
|---|---|---|
| SimpleBench | 41.4% | 17.8% |
| EnigmaEval | 0.9% | 0.8% |
| LiveBench Reasoning | 56.7% | 55.8% |
| LMArena Hard Prompts | 1305 | 1281 |
| DTBench | 67.8% | 64.5% |
| LiveBench Data Analysis | 55% | 60.9% |
| Epoch Capabilities Index | 133.55 | 128.97 |
| ForecastBench | 60.7 | 57.7 |
| LiveBench | 59% | 55.3% |
| ARC-AGI-2 | — | 0% |
| ARC-AGI-1 | — | 4.5% |
| CritPt | — | 0% |
| Chess Puzzles | — | 13% |
| LMCA | — | 16.6% |
Math Claude 3.5 Sonnet leads
Claude 3.5 Sonnet: 18.0 (#282), GPT-4o: 10.2 (#303)
| Benchmark | Claude 3.5 Sonnet | GPT-4o |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 8.5% | 6.4% |
| Omni-MATH | 27.6% | 29.3% |
| LiveBench Math | 52.3% | 49.5% |
| LMArena Math | 1306 | 1285 |
| MATH Level 5 | 56.9% | 53.3% |
| FrontierMath (Feb 2025 set) | 2.1% | 0.3% |
| FrontierMath (Tiers 1-3) | — | 0.4% |
| FrontierMath Tier 4 (v1) | 0% | — |
Knowledge Too close to call
Claude 3.5 Sonnet: 28.5 (#236), GPT-4o: 28.8 (#235)
| Benchmark | Claude 3.5 Sonnet | GPT-4o |
|---|---|---|
| GPQA Diamond | 55.3% | 49.2% |
| Humanity's Last Exam | 4.1% | 2.7% |
| MMLU-Pro | 77.7% | 71.3% |
| Confabulations | 19.9% | 15.3% |
| GPQA (HELM) | 56.5% | 52% |
| LMArena Expert | 1264 | 1250 |
| MMLU | 87.3% | 88.1% |
| SimpleQA Verified | — | 26% |
| Vectara Hallucination Rate | — | 9.6% |
Multimodal GPT-4o leads
Claude 3.5 Sonnet: 26.0 (#115), GPT-4o: 33.9 (#85)
| Benchmark | Claude 3.5 Sonnet | GPT-4o |
|---|---|---|
| LMArena Vision | 1125 | 1137 |
| Video-MME | 60% | 71.9% |
| GeoBench | 62% | 71% |
| VPCT | 33% | 40% |
| ScienceQA | — | 88.5% |
Multilingual Too close to call
Claude 3.5 Sonnet: 43.4 (#180), GPT-4o: 43.4 (#181)
| Benchmark | Claude 3.5 Sonnet | GPT-4o |
|---|---|---|
| LMArena Non-English | 1283 | 1283 |
| LMArena Chinese | 1271 | 1277 |
| LMArena French | 1304 | 1303 |
| LMArena German | 1297 | 1282 |
| LMArena Japanese | 1234 | 1257 |
| LMArena Korean | 1200 | 1234 |
| LMArena Russian | 1305 | 1286 |
| LMArena Spanish | 1289 | 1292 |
Instruction Following Claude 3.5 Sonnet leads
Claude 3.5 Sonnet: 69.0 (#177), GPT-4o: 66.8 (#202)
| Benchmark | Claude 3.5 Sonnet | GPT-4o |
|---|---|---|
| LiveBench Instruction Following | 69.3% | 68.6% |
| IFEval | 85.5% | 81.7% |
| LMArena Instruction Following | 1297 | 1278 |
Long Context Too close to call
Claude 3.5 Sonnet: 39.9 (#162), GPT-4o: 39.4 (#174)
| Benchmark | Claude 3.5 Sonnet | GPT-4o |
|---|---|---|
| LMArena Longer Query | 1311 | 1289 |
| Fiction.LiveBench | — | 66.7% |
Writing & Preference Too close to call
Claude 3.5 Sonnet: 53.0 (#159), GPT-4o: 52.8 (#160)
| Benchmark | Claude 3.5 Sonnet | GPT-4o |
|---|---|---|
| LMArena Text | 1298 | 1300 |
| LMArena Creative Writing | 1292 | 1291 |
| Short-Story Creative Writing | 80.3% | 81.8% |
| WildBench | 79.2% | 82.8% |
| LMArena Multi-Turn | 1326 | 1302 |
| LiveBench Language | 53.8% | 47.6% |
| EQ-Bench Creative Writing | 1451 | — |
Frequently asked questions
Is Claude 3.5 Sonnet better than GPT-4o?
Claude 3.5 Sonnet is the stronger model overall, scoring 34.4 to 28.5 on the Noometry Index.
Is Claude 3.5 Sonnet or GPT-4o better for coding?
Claude 3.5 Sonnet scores higher on coding benchmarks: 39.5 versus 25.1 in the Noometry coding category.
How many benchmarks do Claude 3.5 Sonnet and GPT-4o share?
58 benchmarks have published results for both models. Claude 3.5 Sonnet has 60 scored results on Noometry and GPT-4o has 72.