Model comparison
Claude 3 Opus vs GPT-5 Nano
GPT-5 Nano is the stronger model overall, scoring 33.5 to 29.5 on the Noometry Index.
Last verified . 27 shared benchmarks.
Summary
- They share 27 benchmarks with published results for both. Claude 3 Opus scores higher in 2 categories and GPT-5 Nano in 8 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where GPT-5 Nano leads 29.4 to 14.8.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 4.7% for Claude 3 Opus and 81.1% for GPT-5 Nano.
Side by side
| Claude 3 Opus | GPT-5 Nano | |
|---|---|---|
| Provider | Anthropic | OpenAI |
| Noometry Index | 29.5 | 33.5 |
| Released | 2024-02-29 | 2025-08-07 |
| Weights | Proprietary | Proprietary |
| Context window | — | 400K |
| Max output | — | 128K |
| Input $ / M tokens | — | $0.05 |
| Output $ / M tokens | — | $0.40 |
| Results tracked | 46 | 49 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Too close to call
Claude 3 Opus: 32.9 (#267), GPT-5 Nano: 33.6 (#254)
| Benchmark | Claude 3 Opus | GPT-5 Nano |
|---|---|---|
| WeirdML | 19.2% | 38.1% |
| LMArena Coding | 1264 | 1351 |
| SWE-bench Verified (bash only) | — | 34.8% |
| BigCodeBench Instruct | 45.5% | — |
| LiveBench Coding | 38.6% | — |
| BigCodeBench Complete | 57.4% | — |
| ALE-Bench | — | 718.67 |
| HumanEval+ | 77.4% | — |
| MBPP+ | 73.3% | — |
Agentic & Tool Use GPT-5 Nano leads
Claude 3 Opus: 24.6 (#116), GPT-5 Nano: 25.8 (#106)
| Benchmark | Claude 3 Opus | GPT-5 Nano |
|---|---|---|
| Terminal-Bench | — | 21.8% |
| Berkeley Function Calling Leaderboard | — | 51.5% |
| Cybench | 10% | — |
| METR Time Horizons | 29.5% | — |
Reasoning GPT-5 Nano leads
Claude 3 Opus: 14.6 (#324), GPT-5 Nano: 16.3 (#306)
| Benchmark | Claude 3 Opus | GPT-5 Nano |
|---|---|---|
| Chess Puzzles | 5% | 27% |
| LMArena Hard Prompts | 1245 | 1328 |
| DTBench | 61.6% | 62.7% |
| LMCA | 17% | 7.9% |
| Epoch Capabilities Index | 126.91 | 139.38 |
| ForecastBench | 58.4 | 59.1 |
| ARC-AGI-2 | — | 2.6% |
| SimpleBench | 23.5% | — |
| Kagi LLM Benchmark | — | 62.2% |
| ARC-AGI-1 | — | 20.7% |
| EnigmaEval | 0.8% | — |
| LiveBench Reasoning | 40.6% | — |
| Mystery Game Puzzles | — | 9% |
| LiveBench Data Analysis | 57.9% | — |
| LiveBench | 49.2% | — |
| WinoGrande | 88.5% | — |
Math GPT-5 Nano leads
Claude 3 Opus: 14.8 (#299), GPT-5 Nano: 29.4 (#241)
| Benchmark | Claude 3 Opus | GPT-5 Nano |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 4.7% | 81.1% |
| LMArena Math | 1273 | 1317 |
| MATH Level 5 | 37.5% | 95.2% |
| FrontierMath (Tiers 1-3) | — | 20% |
| FrontierMath Tier 4 | — | 2.4% |
| ProofBench | — | 12% |
| Omni-MATH | — | 54.6% |
| LiveBench Math | 43.6% | — |
| FrontierMath (Feb 2025 set) | — | 8.3% |
| FrontierMath Tier 4 (v1) | — | 2.1% |
Knowledge GPT-5 Nano leads
Claude 3 Opus: 24.5 (#267), GPT-5 Nano: 35.9 (#178)
| Benchmark | Claude 3 Opus | GPT-5 Nano |
|---|---|---|
| GPQA Diamond | 47.2% | 69.4% |
| SimpleQA Verified | 12.6% | 11.7% |
| LMArena Expert | 1223 | 1321 |
| MMLU-Pro | — | 77.8% |
| Confabulations | 22.7% | — |
| Vectara Hallucination Rate | — | 10.5% |
| GPQA (HELM) | — | 67.9% |
| MMLU | 84.6% | — |
Multimodal GPT-5 Nano leads
Claude 3 Opus: 27.1 (#116), GPT-5 Nano: 31.3 (#108)
| Benchmark | Claude 3 Opus | GPT-5 Nano |
|---|---|---|
| LMArena Vision | 1023 | 1159 |
| VPCT | — | 37.2% |
Multilingual GPT-5 Nano leads
Claude 3 Opus: 41.4 (#207), GPT-5 Nano: 45.3 (#172)
| Benchmark | Claude 3 Opus | GPT-5 Nano |
|---|---|---|
| LMArena Non-English | 1258 | 1313 |
| LMArena Chinese | 1248 | 1356 |
| LMArena German | 1258 | 1327 |
| LMArena Japanese | 1204 | 1226 |
| LMArena Korean | 1187 | 1269 |
| LMArena Russian | 1280 | 1296 |
| LMArena Spanish | 1246 | 1360 |
| LMArena French | 1275 | — |
Instruction Following GPT-5 Nano leads
Claude 3 Opus: 64.1 (#228), GPT-5 Nano: 75.0 (#79)
| Benchmark | Claude 3 Opus | GPT-5 Nano |
|---|---|---|
| LMArena Instruction Following | 1248 | 1306 |
| LiveBench Instruction Following | 63.9% | — |
| IFEval | — | 93.2% |
Long Context Claude 3 Opus leads
Claude 3 Opus: 38.2 (#202), GPT-5 Nano: 31.3 (#281)
| Benchmark | Claude 3 Opus | GPT-5 Nano |
|---|---|---|
| LMArena Longer Query | 1259 | 1312 |
| Fiction.LiveBench | — | 44.4% |
Writing & Preference Claude 3 Opus leads
Claude 3 Opus: 47.2 (#213), GPT-5 Nano: 39.1 (#249)
| Benchmark | Claude 3 Opus | GPT-5 Nano |
|---|---|---|
| LMArena Text | 1262 | 1320 |
| LMArena Creative Writing | 1235 | 1249 |
| LMArena Multi-Turn | 1275 | 1311 |
| EQ-Bench Creative Writing | — | 705 |
| WildBench | — | 80.6% |
| LiveBench Language | 50.4% | — |
Frequently asked questions
Is Claude 3 Opus better than GPT-5 Nano?
GPT-5 Nano is the stronger model overall, scoring 33.5 to 29.5 on the Noometry Index.
Is Claude 3 Opus or GPT-5 Nano better for coding?
They score almost the same on coding (32.9 vs 33.6); test both on your own repository before choosing.
How many benchmarks do Claude 3 Opus and GPT-5 Nano share?
27 benchmarks have published results for both models. Claude 3 Opus has 46 scored results on Noometry and GPT-5 Nano has 49.