Model comparison
Claude 3 Opus vs GPT-5.4 nano
GPT-5.4 nano is the stronger model overall, scoring 41.9 to 29.3 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 0 categories and GPT-5.4 nano in 9 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where GPT-5.4 nano leads 40.7 to 13.9.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 4.7% for Claude 3 Opus and 87.8% for GPT-5.4 nano.
Side by side
| Claude 3 Opus | GPT-5.4 nano | |
|---|---|---|
| Provider | Anthropic | OpenAI |
| Noometry Index | 29.3 | 41.9 |
| Released | 2024-02-29 | 2026-03-17 |
| Weights | Proprietary | Proprietary |
| Context window | — | 400K |
| Max output | — | 128K |
| Input $ / M tokens | — | $0.20 |
| Output $ / M tokens | — | $1.25 |
| Results tracked | 46 | 40 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-5.4 nano leads
Claude 3 Opus: 32.4 (#261), GPT-5.4 nano: 43.1 (#82)
| Benchmark | Claude 3 Opus | GPT-5.4 nano |
|---|---|---|
| WeirdML | 19.2% | 49.2% |
| LMArena Coding | 1264 | 1404 |
| SciCode | — | 46.9% |
| BigCodeBench Instruct | 45.5% | — |
| LiveBench Coding | 38.6% | — |
| BigCodeBench Complete | 57.4% | — |
| ALE-Bench | — | 1,005 |
| HumanEval+ | 77.4% | — |
| MBPP+ | 73.3% | — |
Agentic & Tool Use Not comparable
Claude 3 Opus: 24.3 (#110), GPT-5.4 nano: —
| Benchmark | Claude 3 Opus | GPT-5.4 nano |
|---|---|---|
| Cybench | 10% | — |
| METR Time Horizons | 29.5% | — |
Reasoning GPT-5.4 nano leads
Claude 3 Opus: 14.2 (#318), GPT-5.4 nano: 23.8 (#159)
| Benchmark | Claude 3 Opus | GPT-5.4 nano |
|---|---|---|
| Chess Puzzles | 5% | 30% |
| LMArena Hard Prompts | 1246 | 1380 |
| DTBench | 61.6% | 80.3% |
| LMCA | 17% | 36.9% |
| Epoch Capabilities Index | 126.91 | 145.78 |
| ForecastBench | 58.4 | 57.3 |
| ARC-AGI-2 | — | 5.7% |
| SimpleBench | 23.5% | — |
| Kagi LLM Benchmark | — | 39.7% |
| ARC-AGI-1 | — | 51.5% |
| CritPt | — | 9.3% |
| EnigmaEval | 0.8% | — |
| LiveBench Reasoning | 40.6% | — |
| Mystery Game Puzzles | — | 9% |
| LiveBench Data Analysis | 57.9% | — |
| LiveBench | 49.2% | — |
| WinoGrande | 88.5% | — |
Math GPT-5.4 nano leads
Claude 3 Opus: 13.9 (#293), GPT-5.4 nano: 40.7 (#76)
| Benchmark | Claude 3 Opus | GPT-5.4 nano |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 4.7% | 87.8% |
| LMArena Math | 1273 | 1406 |
| FrontierMath (Tiers 1-3) | — | 44.9% |
| FrontierMath Tier 4 | — | 12.2% |
| ProofBench | — | 5% |
| LiveBench Math | 43.6% | — |
| MATH Level 5 | 37.5% | — |
| FrontierMath (Feb 2025 set) | — | 25.9% |
| FrontierMath Tier 4 (v1) | — | 6.3% |
Knowledge GPT-5.4 nano leads
Claude 3 Opus: 24.5 (#262), GPT-5.4 nano: 41.8 (#94)
| Benchmark | Claude 3 Opus | GPT-5.4 nano |
|---|---|---|
| GPQA Diamond | 47.2% | 78.5% |
| SimpleQA Verified | 12.6% | 11.7% |
| LMArena Expert | 1223 | 1397 |
| Confabulations | 22.7% | — |
| Vectara Hallucination Rate | — | 3.1% |
| MMLU | 84.6% | — |
Multimodal GPT-5.4 nano leads
Claude 3 Opus: 26.6 (#112), GPT-5.4 nano: 36.0 (#73)
| Benchmark | Claude 3 Opus | GPT-5.4 nano |
|---|---|---|
| LMArena Vision | 1023 | 1196 |
Multilingual GPT-5.4 nano leads
Claude 3 Opus: 41.6 (#202), GPT-5.4 nano: 48.8 (#136)
| Benchmark | Claude 3 Opus | GPT-5.4 nano |
|---|---|---|
| LMArena Non-English | 1258 | 1359 |
| LMArena Chinese | 1248 | 1391 |
| LMArena French | 1274 | 1396 |
| LMArena German | 1258 | 1368 |
| LMArena Japanese | 1204 | 1342 |
| LMArena Korean | 1187 | 1321 |
| LMArena Russian | 1280 | 1363 |
| LMArena Spanish | 1246 | 1368 |
Instruction Following GPT-5.4 nano leads
Claude 3 Opus: 64.3 (#223), GPT-5.4 nano: 72.1 (#139)
| Benchmark | Claude 3 Opus | GPT-5.4 nano |
|---|---|---|
| LMArena Instruction Following | 1248 | 1362 |
| LiveBench Instruction Following | 63.9% | — |
Long Context GPT-5.4 nano leads
Claude 3 Opus: 38.2 (#197), GPT-5.4 nano: 41.6 (#131)
| Benchmark | Claude 3 Opus | GPT-5.4 nano |
|---|---|---|
| LMArena Longer Query | 1259 | 1365 |
Writing & Preference GPT-5.4 nano leads
Claude 3 Opus: 47.4 (#208), GPT-5.4 nano: 55.9 (#137)
| Benchmark | Claude 3 Opus | GPT-5.4 nano |
|---|---|---|
| LMArena Text | 1262 | 1372 |
| LMArena Creative Writing | 1235 | 1313 |
| LMArena Multi-Turn | 1275 | 1382 |
| LiveBench Language | 50.4% | — |
Frequently asked questions
Is Claude 3 Opus better than GPT-5.4 nano?
GPT-5.4 nano is the stronger model overall, scoring 41.9 to 29.3 on the Noometry Index.
Is Claude 3 Opus or GPT-5.4 nano better for coding?
GPT-5.4 nano scores higher on coding benchmarks: 43.1 versus 32.4 in the Noometry coding category.
How many benchmarks do Claude 3 Opus and GPT-5.4 nano share?
27 benchmarks have published results for both models. Claude 3 Opus has 46 scored results on Noometry and GPT-5.4 nano has 40.