Model comparison
Claude 3 Opus vs GPT-5 Mini
GPT-5 Mini is the stronger model overall, scoring 41.8 to 29.5 on the Noometry Index.
Last verified . 30 shared benchmarks.
Summary
- They share 30 benchmarks with published results for both. Claude 3 Opus scores higher in 0 categories and GPT-5 Mini in 10 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in math, where GPT-5 Mini leads 46.7 to 14.8.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 4.7% for Claude 3 Opus and 86.7% for GPT-5 Mini.
Side by side
| Claude 3 Opus | GPT-5 Mini | |
|---|---|---|
| Provider | Anthropic | OpenAI |
| Noometry Index | 29.5 | 41.8 |
| Released | 2024-02-29 | 2025-08-07 |
| Weights | Proprietary | Proprietary |
| Context window | — | 400K |
| Max output | — | 128K |
| Input $ / M tokens | — | $0.25 |
| Output $ / M tokens | — | $2 |
| Results tracked | 46 | 60 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-5 Mini leads
Claude 3 Opus: 32.9 (#267), GPT-5 Mini: 40.1 (#146)
| Benchmark | Claude 3 Opus | GPT-5 Mini |
|---|---|---|
| WeirdML | 19.2% | 52.7% |
| LMArena Coding | 1264 | 1406 |
| SWE-bench Verified | — | 64.7% |
| SWE-bench Verified (bash only) | — | 59.8% |
| SWE-bench Multilingual | — | 39.7% |
| SciCode | — | 39.2% |
| BigCodeBench Instruct | 45.5% | — |
| LiveBench Coding | 38.6% | — |
| BigCodeBench Complete | 57.4% | — |
| ALE-Bench | — | 799.77 |
| AlgoTune | — | 1.38 |
| HumanEval+ | 77.4% | — |
| MBPP+ | 73.3% | — |
Agentic & Tool Use GPT-5 Mini leads
Claude 3 Opus: 24.6 (#116), GPT-5 Mini: 31.1 (#70)
| Benchmark | Claude 3 Opus | GPT-5 Mini |
|---|---|---|
| Terminal-Bench | — | 34.8% |
| Berkeley Function Calling Leaderboard | — | 55.5% |
| Cybench | 10% | — |
| METR Time Horizons | 29.5% | — |
| Vending-Bench 2 | — | -31.18 |
Reasoning GPT-5 Mini leads
Claude 3 Opus: 14.6 (#324), GPT-5 Mini: 23.9 (#168)
| Benchmark | Claude 3 Opus | GPT-5 Mini |
|---|---|---|
| Chess Puzzles | 5% | 30% |
| EnigmaEval | 0.8% | 8.2% |
| LMArena Hard Prompts | 1245 | 1380 |
| DTBench | 61.6% | 80.5% |
| LMCA | 17% | 34.2% |
| Epoch Capabilities Index | 126.91 | 145.52 |
| ForecastBench | 58.4 | 61 |
| ARC-AGI-2 | — | 4.4% |
| SimpleBench | 23.5% | — |
| Kagi LLM Benchmark | — | 70.3% |
| ARC-AGI-1 | — | 54.3% |
| CritPt | — | 0% |
| LiveBench Reasoning | 40.6% | — |
| Mystery Game Puzzles | — | 10% |
| LiveBench Data Analysis | 57.9% | — |
| LiveBench | 49.2% | — |
| WinoGrande | 88.5% | — |
Math GPT-5 Mini leads
Claude 3 Opus: 14.8 (#299), GPT-5 Mini: 46.7 (#69)
| Benchmark | Claude 3 Opus | GPT-5 Mini |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 4.7% | 86.7% |
| LMArena Math | 1273 | 1378 |
| MATH Level 5 | 37.5% | 97.8% |
| FrontierMath (Tiers 1-3) | — | 46.7% |
| FrontierMath Tier 4 | — | 12.2% |
| ProofBench | — | 9% |
| Omni-MATH | — | 72.2% |
| LiveBench Math | 43.6% | — |
| FrontierMath (Feb 2025 set) | — | 27.2% |
| FrontierMath Tier 4 (v1) | — | 6.3% |
Knowledge GPT-5 Mini leads
Claude 3 Opus: 24.5 (#267), GPT-5 Mini: 45.6 (#86)
| Benchmark | Claude 3 Opus | GPT-5 Mini |
|---|---|---|
| GPQA Diamond | 47.2% | 75% |
| SimpleQA Verified | 12.6% | 21.6% |
| Confabulations | 22.7% | 13.3% |
| LMArena Expert | 1223 | 1379 |
| Humanity's Last Exam | — | 19.4% |
| MMLU-Pro | — | 83.5% |
| Vectara Hallucination Rate | — | 12.9% |
| GPQA (HELM) | — | 75.6% |
| MMLU | 84.6% | — |
Multimodal GPT-5 Mini leads
Claude 3 Opus: 27.1 (#116), GPT-5 Mini: 35.6 (#85)
| Benchmark | Claude 3 Opus | GPT-5 Mini |
|---|---|---|
| LMArena Vision | 1023 | 1202 |
| VPCT | — | 40.2% |
Multilingual GPT-5 Mini leads
Claude 3 Opus: 41.4 (#207), GPT-5 Mini: 48.9 (#137)
| Benchmark | Claude 3 Opus | GPT-5 Mini |
|---|---|---|
| LMArena Non-English | 1258 | 1363 |
| LMArena Chinese | 1248 | 1385 |
| LMArena French | 1275 | 1386 |
| LMArena German | 1258 | 1366 |
| LMArena Japanese | 1204 | 1341 |
| LMArena Korean | 1187 | 1308 |
| LMArena Russian | 1280 | 1362 |
| LMArena Spanish | 1246 | 1355 |
Instruction Following GPT-5 Mini leads
Claude 3 Opus: 64.1 (#228), GPT-5 Mini: 76.2 (#46)
| Benchmark | Claude 3 Opus | GPT-5 Mini |
|---|---|---|
| LMArena Instruction Following | 1248 | 1357 |
| LiveBench Instruction Following | 63.9% | — |
| IFEval | — | 92.7% |
Long Context GPT-5 Mini leads
Claude 3 Opus: 38.2 (#202), GPT-5 Mini: 41.9 (#132)
| Benchmark | Claude 3 Opus | GPT-5 Mini |
|---|---|---|
| LMArena Longer Query | 1259 | 1355 |
| Fiction.LiveBench | — | 69.4% |
Writing & Preference GPT-5 Mini leads
Claude 3 Opus: 47.2 (#213), GPT-5 Mini: 55.2 (#148)
| Benchmark | Claude 3 Opus | GPT-5 Mini |
|---|---|---|
| LMArena Text | 1262 | 1373 |
| LMArena Creative Writing | 1235 | 1325 |
| LMArena Multi-Turn | 1275 | 1363 |
| Short-Story Creative Writing | — | 83.1% |
| EQ-Bench Creative Writing | — | 1313 |
| WildBench | — | 85.5% |
| LiveBench Language | 50.4% | — |
Frequently asked questions
Is Claude 3 Opus better than GPT-5 Mini?
GPT-5 Mini is the stronger model overall, scoring 41.8 to 29.5 on the Noometry Index.
Is Claude 3 Opus or GPT-5 Mini better for coding?
GPT-5 Mini scores higher on coding benchmarks: 40.1 versus 32.9 in the Noometry coding category.
How many benchmarks do Claude 3 Opus and GPT-5 Mini share?
30 benchmarks have published results for both models. Claude 3 Opus has 46 scored results on Noometry and GPT-5 Mini has 60.