Model comparison
Claude 3 Opus vs GPT-4o mini
Claude 3 Opus is the stronger model overall, scoring 29.3 to 25.4 on the Noometry Index.
Last verified . 41 shared benchmarks.
Summary
- They share 41 benchmarks with published results for both. Claude 3 Opus scores higher in 7 categories and GPT-4o mini in 3 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in coding, where Claude 3 Opus leads 32.4 to 21.6.
- The biggest single-benchmark swing is LiveBench Language: 50.4% for Claude 3 Opus and 28.6% for GPT-4o mini.
Side by side
| Claude 3 Opus | GPT-4o mini | |
|---|---|---|
| Provider | Anthropic | OpenAI |
| Noometry Index | 29.3 | 25.4 |
| Released | 2024-02-29 | 2024-07-18 |
| Weights | Proprietary | Proprietary |
| Context window | — | 128K |
| Max output | — | 16K |
| Input $ / M tokens | — | $0.15 |
| Output $ / M tokens | — | $0.60 |
| Results tracked | 46 | 60 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Claude 3 Opus leads
Claude 3 Opus: 32.4 (#261), GPT-4o mini: 21.6 (#330)
| Benchmark | Claude 3 Opus | GPT-4o mini |
|---|---|---|
| WeirdML | 19.2% | 11.8% |
| BigCodeBench Instruct | 45.5% | 46.1% |
| LiveBench Coding | 38.6% | 43.1% |
| LMArena Coding | 1264 | 1290 |
| BigCodeBench Complete | 57.4% | 57.4% |
| HumanEval+ | 77.4% | 83.5% |
| MBPP+ | 73.3% | 72.2% |
| Aider Polyglot | — | 3.6% |
Agentic & Tool Use GPT-4o mini leads
Claude 3 Opus: 24.3 (#110), GPT-4o mini: 27.2 (#95)
| Benchmark | Claude 3 Opus | GPT-4o mini |
|---|---|---|
| Cybench | 10% | — |
| BALROG | — | 17.4% |
| METR Time Horizons | 29.5% | — |
Reasoning Claude 3 Opus leads
Claude 3 Opus: 14.2 (#318), GPT-4o mini: 8.7 (#343)
| Benchmark | Claude 3 Opus | GPT-4o mini |
|---|---|---|
| SimpleBench | 23.5% | 10.7% |
| Chess Puzzles | 5% | 0% |
| LiveBench Reasoning | 40.6% | 32.8% |
| LMArena Hard Prompts | 1246 | 1267 |
| DTBench | 61.6% | 54.4% |
| LiveBench Data Analysis | 57.9% | 50% |
| LMCA | 17% | 10.4% |
| Epoch Capabilities Index | 126.91 | 126.56 |
| LiveBench | 49.2% | 41.3% |
| ARC-AGI-2 | — | 0% |
| Kagi LLM Benchmark | — | 28.8% |
| EnigmaEval | 0.8% | — |
| Mystery Game Puzzles | — | 12% |
| ForecastBench | 58.4 | — |
| PIQA | — | 88.7% |
| WinoGrande | 88.5% | — |
Math Claude 3 Opus leads
Claude 3 Opus: 13.9 (#293), GPT-4o mini: 9.9 (#307)
| Benchmark | Claude 3 Opus | GPT-4o mini |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 4.7% | 6.9% |
| LiveBench Math | 43.6% | 36.3% |
| LMArena Math | 1273 | 1267 |
| MATH Level 5 | 37.5% | 52.6% |
| FrontierMath (Tiers 1-3) | — | 0.7% |
| Omni-MATH | — | 28% |
| GSM8K | — | 91.3% |
Knowledge Claude 3 Opus leads
Claude 3 Opus: 24.5 (#262), GPT-4o mini: 17.7 (#279)
| Benchmark | Claude 3 Opus | GPT-4o mini |
|---|---|---|
| GPQA Diamond | 47.2% | 37.7% |
| SimpleQA Verified | 12.6% | 8.3% |
| Confabulations | 22.7% | 37.2% |
| LMArena Expert | 1223 | 1235 |
| MMLU | 84.6% | 81.8% |
| MMLU-Pro | — | 60.3% |
| GPQA (HELM) | — | 36.8% |
| BoolQ | — | 88.7% |
Multimodal Claude 3 Opus leads
Claude 3 Opus: 26.6 (#112), GPT-4o mini: 25.3 (#117)
| Benchmark | Claude 3 Opus | GPT-4o mini |
|---|---|---|
| LMArena Vision | 1023 | 1066 |
| Video-MME | — | 64.8% |
| GeoBench | — | 64% |
| VPCT | — | 34% |
Multilingual Too close to call
Claude 3 Opus: 41.6 (#202), GPT-4o mini: 42.2 (#194)
| Benchmark | Claude 3 Opus | GPT-4o mini |
|---|---|---|
| LMArena Non-English | 1258 | 1266 |
| LMArena Chinese | 1248 | 1264 |
| LMArena French | 1274 | 1296 |
| LMArena German | 1258 | 1272 |
| LMArena Japanese | 1204 | 1216 |
| LMArena Korean | 1187 | 1195 |
| LMArena Russian | 1280 | 1275 |
| LMArena Spanish | 1246 | 1276 |
Instruction Following Claude 3 Opus leads
Claude 3 Opus: 64.3 (#223), GPT-4o mini: 62.1 (#234)
| Benchmark | Claude 3 Opus | GPT-4o mini |
|---|---|---|
| LiveBench Instruction Following | 63.9% | 56.8% |
| LMArena Instruction Following | 1248 | 1259 |
| IFEval | — | 78.2% |
Long Context Too close to call
Claude 3 Opus: 38.2 (#197), GPT-4o mini: 39.1 (#181)
| Benchmark | Claude 3 Opus | GPT-4o mini |
|---|---|---|
| LMArena Longer Query | 1259 | 1289 |
Writing & Preference Claude 3 Opus leads
Claude 3 Opus: 47.4 (#208), GPT-4o mini: 39.7 (#243)
| Benchmark | Claude 3 Opus | GPT-4o mini |
|---|---|---|
| LMArena Text | 1262 | 1286 |
| LMArena Creative Writing | 1235 | 1268 |
| LMArena Multi-Turn | 1275 | 1285 |
| LiveBench Language | 50.4% | 28.6% |
| Short-Story Creative Writing | — | 67.2% |
| EQ-Bench Creative Writing | — | 873 |
| WildBench | — | 79.1% |
Frequently asked questions
Is Claude 3 Opus better than GPT-4o mini?
Claude 3 Opus is the stronger model overall, scoring 29.3 to 25.4 on the Noometry Index.
Is Claude 3 Opus or GPT-4o mini better for coding?
Claude 3 Opus scores higher on coding benchmarks: 32.4 versus 21.6 in the Noometry coding category.
How many benchmarks do Claude 3 Opus and GPT-4o mini share?
41 benchmarks have published results for both models. Claude 3 Opus has 46 scored results on Noometry and GPT-4o mini has 60.