Model comparison
Claude Opus 4 vs GPT-4.5
Claude Opus 4 is the stronger model overall, scoring 43.1 to 37.2 on the Noometry Index.
Last verified . 35 shared benchmarks.
Summary
- They share 35 benchmarks with published results for both. Claude Opus 4 scores higher in 7 categories and GPT-4.5 in 3 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Claude Opus 4 leads 27.3 to 13.9.
- The biggest single-benchmark swing is Aider Polyglot: 72% for Claude Opus 4 and 44.9% for GPT-4.5.
Side by side
| Claude Opus 4 | GPT-4.5 | |
|---|---|---|
| Provider | Anthropic | OpenAI |
| Noometry Index | 43.1 | 37.2 |
| Released | 2025-05-22 | 2025-02-27 |
| Weights | Proprietary | Proprietary |
| Context window | 200K | — |
| Max output | 32K | — |
| Input $ / M tokens | $15 | — |
| Output $ / M tokens | $75 | — |
| Results tracked | 56 | 42 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Claude Opus 4 leads
Claude Opus 4: 47.2 (#62), GPT-4.5: 42.2 (#109)
| Benchmark | Claude Opus 4 | GPT-4.5 |
|---|---|---|
| Aider Polyglot | 72% | 44.9% |
| WeirdML | 43.7% | 39.4% |
| LMArena Coding | 1442 | 1396 |
| SWE-bench Verified | 70.7% | — |
| SWE-bench Verified (bash only) | 67.6% | — |
| GSO | 6.9% | — |
| LiveBench Coding | — | 75.2% |
| AlgoTune | 1.33 | — |
Agentic & Tool Use Claude Opus 4 leads
Claude Opus 4: 34.8 (#42), GPT-4.5: 27.9 (#97)
| Benchmark | Claude Opus 4 | GPT-4.5 |
|---|---|---|
| Cybench | 38% | 17.5% |
| DeepResearch Bench | 46.8% | — |
| LMArena Search | 1127 | — |
| METR Time Horizons | 63.9% | — |
Reasoning Claude Opus 4 leads
Claude Opus 4: 27.3 (#121), GPT-4.5: 13.9 (#330)
| Benchmark | Claude Opus 4 | GPT-4.5 |
|---|---|---|
| ARC-AGI-2 | 8.6% | 0.8% |
| SimpleBench | 58.8% | 34.5% |
| ARC-AGI-1 | 35.7% | 10.3% |
| EnigmaEval | 5.6% | 3.2% |
| LMArena Hard Prompts | 1399 | 1403 |
| Epoch Capabilities Index | 142.67 | 136.74 |
| ForecastBench | 61.1 | 61.7 |
| Kagi LLM Benchmark | 74.3% | — |
| CritPt | 0.3% | — |
| LiveBench Reasoning | — | 71.1% |
| DTBench | 81.6% | — |
| LiveBench Data Analysis | — | 64.3% |
| LMCA | 37.4% | — |
| LiveBench | — | 69% |
Math Claude Opus 4 leads
Claude Opus 4: 42.0 (#86), GPT-4.5: 32.6 (#211)
| Benchmark | Claude Opus 4 | GPT-4.5 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 64.4% | 37.8% |
| LMArena Math | 1390 | 1412 |
| MATH Level 5 | 85% | 78.6% |
| Omni-MATH | 61.6% | — |
| LiveBench Math | — | 69.3% |
| FrontierMath (Feb 2025 set) | 4.5% | — |
| FrontierMath Tier 4 (v1) | 4.2% | — |
Knowledge Claude Opus 4 leads
Claude Opus 4: 44.0 (#88), GPT-4.5: 32.5 (#211)
| Benchmark | Claude Opus 4 | GPT-4.5 |
|---|---|---|
| GPQA Diamond | 76.3% | 68.7% |
| Humanity's Last Exam | 10.7% | 5.4% |
| Confabulations | 15.9% | 13.6% |
| LMArena Expert | 1386 | 1394 |
| MMLU-Pro | 87.5% | — |
| Vectara Hallucination Rate | 12% | — |
| GPQA (HELM) | 70.8% | — |
Multimodal GPT-4.5 leads
Claude Opus 4: 31.5 (#106), GPT-4.5: 37.6 (#71)
| Benchmark | Claude Opus 4 | GPT-4.5 |
|---|---|---|
| LMArena Vision | 1192 | 1195 |
| VPCT | 38% | 45% |
| GeoBench | 49% | — |
Multilingual GPT-4.5 leads
Claude Opus 4: 48.8 (#138), GPT-4.5: 52.5 (#83)
| Benchmark | Claude Opus 4 | GPT-4.5 |
|---|---|---|
| LMArena Non-English | 1362 | 1413 |
| LMArena Chinese | 1386 | 1421 |
| LMArena French | 1372 | 1418 |
| LMArena German | 1391 | 1457 |
| LMArena Japanese | 1331 | 1416 |
| LMArena Korean | 1321 | 1392 |
| LMArena Russian | 1392 | 1419 |
| LMArena Spanish | 1389 | — |
Instruction Following Claude Opus 4 leads
Claude Opus 4: 77.1 (#28), GPT-4.5: 72.6 (#134)
| Benchmark | Claude Opus 4 | GPT-4.5 |
|---|---|---|
| LMArena Instruction Following | 1406 | 1404 |
| LiveBench Instruction Following | — | 72.3% |
| IFEval | 91.8% | — |
Long Context Too close to call
Claude Opus 4: 39.6 (#172), GPT-4.5: 40.4 (#155)
| Benchmark | Claude Opus 4 | GPT-4.5 |
|---|---|---|
| Fiction.LiveBench | 61.1% | 63.9% |
| LMArena Longer Query | 1422 | 1406 |
Writing & Preference Claude Opus 4 leads
Claude Opus 4: 61.2 (#89), GPT-4.5: 56.9 (#134)
| Benchmark | Claude Opus 4 | GPT-4.5 |
|---|---|---|
| LMArena Text | 1377 | 1417 |
| LMArena Creative Writing | 1387 | 1394 |
| Short-Story Creative Writing | 83.6% | 75.6% |
| EQ-Bench Creative Writing | 1580 | 1258 |
| LMArena Multi-Turn | 1396 | 1444 |
| WildBench | 85.2% | — |
| LiveBench Language | — | 61.5% |
Frequently asked questions
Is Claude Opus 4 better than GPT-4.5?
Claude Opus 4 is the stronger model overall, scoring 43.1 to 37.2 on the Noometry Index.
Is Claude Opus 4 or GPT-4.5 better for coding?
Claude Opus 4 scores higher on coding benchmarks: 47.2 versus 42.2 in the Noometry coding category.
How many benchmarks do Claude Opus 4 and GPT-4.5 share?
35 benchmarks have published results for both models. Claude Opus 4 has 56 scored results on Noometry and GPT-4.5 has 42.