Model comparison
Claude 3.5 Sonnet vs GPT-4o mini
Claude 3.5 Sonnet is the stronger model overall, scoring 34.6 to 25.5 on the Noometry Index.
Last verified . 50 shared benchmarks.
Summary
- They share 50 benchmarks with published results for both. Claude 3.5 Sonnet scores higher in 10 categories and GPT-4o mini in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in coding, where Claude 3.5 Sonnet leads 39.0 to 22.0.
- The biggest single-benchmark swing is Aider Polyglot: 51.6% for Claude 3.5 Sonnet and 3.6% for GPT-4o mini.
Side by side
| Claude 3.5 Sonnet | GPT-4o mini | |
|---|---|---|
| Provider | Anthropic | OpenAI |
| Noometry Index | 34.6 | 25.5 |
| Released | 2024-06-20 | 2024-07-18 |
| Weights | Proprietary | Proprietary |
| Context window | — | 128K |
| Max output | — | 16K |
| Input $ / M tokens | — | $0.15 |
| Output $ / M tokens | — | $0.60 |
| Results tracked | 60 | 60 |
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: 39.0 (#165), GPT-4o mini: 22.0 (#335)
| Benchmark | Claude 3.5 Sonnet | GPT-4o mini |
|---|---|---|
| Aider Polyglot | 51.6% | 3.6% |
| WeirdML | 40% | 11.8% |
| BigCodeBench Instruct | 46.8% | 46.1% |
| LiveBench Coding | 67.1% | 43.1% |
| LMArena Coding | 1342 | 1290 |
| BigCodeBench Complete | 58.6% | 57.4% |
| HumanEval+ | 81.7% | 83.5% |
| MBPP+ | 74.3% | 72.2% |
| GSO | 4.6% | — |
| CadEval | 48% | — |
Agentic & Tool Use Claude 3.5 Sonnet leads
Claude 3.5 Sonnet: 32.3 (#67), GPT-4o mini: 27.5 (#101)
| Benchmark | Claude 3.5 Sonnet | GPT-4o mini |
|---|---|---|
| BALROG | 32.6% | 17.4% |
| TheAgentCompany | 24% | — |
| Cybench | 17.5% | — |
| METR Time Horizons | 45.2% | — |
Reasoning Claude 3.5 Sonnet leads
Claude 3.5 Sonnet: 23.1 (#183), GPT-4o mini: 8.7 (#347)
| Benchmark | Claude 3.5 Sonnet | GPT-4o mini |
|---|---|---|
| SimpleBench | 41.4% | 10.7% |
| LiveBench Reasoning | 56.7% | 32.8% |
| LMArena Hard Prompts | 1305 | 1267 |
| DTBench | 67.8% | 54.4% |
| LiveBench Data Analysis | 55% | 50% |
| Epoch Capabilities Index | 133.55 | 126.56 |
| LiveBench | 59% | 41.3% |
| ARC-AGI-2 | — | 0% |
| Kagi LLM Benchmark | — | 28.8% |
| Chess Puzzles | — | 0% |
| EnigmaEval | 0.9% | — |
| Mystery Game Puzzles | — | 12% |
| LMCA | — | 10.4% |
| ForecastBench | 60.7 | — |
| PIQA | — | 88.7% |
Math Claude 3.5 Sonnet leads
Claude 3.5 Sonnet: 19.2 (#288), GPT-4o mini: 10.4 (#314)
| Benchmark | Claude 3.5 Sonnet | GPT-4o mini |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 8.5% | 6.9% |
| Omni-MATH | 27.6% | 28% |
| LiveBench Math | 52.3% | 36.3% |
| LMArena Math | 1307 | 1267 |
| MATH Level 5 | 56.9% | 52.6% |
| FrontierMath (Tiers 1-3) | — | 0.7% |
| FrontierMath (Feb 2025 set) | 2.1% | — |
| FrontierMath Tier 4 (v1) | 0% | — |
| GSM8K | — | 91.3% |
Knowledge Claude 3.5 Sonnet leads
Claude 3.5 Sonnet: 28.6 (#245), GPT-4o mini: 17.7 (#284)
| Benchmark | Claude 3.5 Sonnet | GPT-4o mini |
|---|---|---|
| GPQA Diamond | 55.3% | 37.7% |
| MMLU-Pro | 77.7% | 60.3% |
| Confabulations | 19.9% | 37.2% |
| GPQA (HELM) | 56.5% | 36.8% |
| LMArena Expert | 1265 | 1235 |
| MMLU | 87.3% | 81.8% |
| Humanity's Last Exam | 4.1% | — |
| SimpleQA Verified | — | 8.3% |
| BoolQ | — | 88.7% |
Multimodal Too close to call
Claude 3.5 Sonnet: 26.5 (#120), GPT-4o mini: 25.9 (#122)
| Benchmark | Claude 3.5 Sonnet | GPT-4o mini |
|---|---|---|
| LMArena Vision | 1125 | 1066 |
| Video-MME | 60% | 64.8% |
| GeoBench | 62% | 64% |
| VPCT | 33% | 34% |
Multilingual Claude 3.5 Sonnet leads
Claude 3.5 Sonnet: 43.2 (#185), GPT-4o mini: 42.0 (#199)
| Benchmark | Claude 3.5 Sonnet | GPT-4o mini |
|---|---|---|
| LMArena Non-English | 1283 | 1266 |
| LMArena Chinese | 1272 | 1265 |
| LMArena French | 1305 | 1297 |
| LMArena German | 1297 | 1272 |
| LMArena Japanese | 1234 | 1216 |
| LMArena Korean | 1200 | 1195 |
| LMArena Russian | 1306 | 1275 |
| LMArena Spanish | 1290 | 1276 |
Instruction Following Claude 3.5 Sonnet leads
Claude 3.5 Sonnet: 68.8 (#182), GPT-4o mini: 61.9 (#239)
| Benchmark | Claude 3.5 Sonnet | GPT-4o mini |
|---|---|---|
| LiveBench Instruction Following | 69.3% | 56.8% |
| IFEval | 85.5% | 78.2% |
| LMArena Instruction Following | 1297 | 1258 |
Long Context Too close to call
Claude 3.5 Sonnet: 39.9 (#167), GPT-4o mini: 39.1 (#186)
| Benchmark | Claude 3.5 Sonnet | GPT-4o mini |
|---|---|---|
| LMArena Longer Query | 1311 | 1289 |
Writing & Preference Claude 3.5 Sonnet leads
Claude 3.5 Sonnet: 52.9 (#164), GPT-4o mini: 39.5 (#248)
| Benchmark | Claude 3.5 Sonnet | GPT-4o mini |
|---|---|---|
| LMArena Text | 1298 | 1286 |
| LMArena Creative Writing | 1292 | 1268 |
| Short-Story Creative Writing | 80.3% | 67.2% |
| EQ-Bench Creative Writing | 1451 | 873 |
| WildBench | 79.2% | 79.1% |
| LMArena Multi-Turn | 1326 | 1285 |
| LiveBench Language | 53.8% | 28.6% |
Frequently asked questions
Is Claude 3.5 Sonnet better than GPT-4o mini?
Claude 3.5 Sonnet is the stronger model overall, scoring 34.6 to 25.5 on the Noometry Index.
Is Claude 3.5 Sonnet or GPT-4o mini better for coding?
Claude 3.5 Sonnet scores higher on coding benchmarks: 39.0 versus 22.0 in the Noometry coding category.
How many benchmarks do Claude 3.5 Sonnet and GPT-4o mini share?
50 benchmarks have published results for both models. Claude 3.5 Sonnet has 60 scored results on Noometry and GPT-4o mini has 60.