Model comparison
Claude 3 Opus vs Claude Opus 4
Claude Opus 4 is the stronger model overall, scoring 43.1 to 29.5 on the Noometry Index.
Last verified . 31 shared benchmarks.
Summary
- They share 31 benchmarks with published results for both. Claude 3 Opus scores higher in 0 categories and Claude Opus 4 in 10 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in math, where Claude Opus 4 leads 42.0 to 14.8.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 4.7% for Claude 3 Opus and 64.4% for Claude Opus 4.
Side by side
| Claude 3 Opus | Claude Opus 4 | |
|---|---|---|
| Provider | Anthropic | Anthropic |
| Noometry Index | 29.5 | 43.1 |
| Released | 2024-02-29 | 2025-05-22 |
| Weights | Proprietary | Proprietary |
| Context window | — | 200K |
| Max output | — | 32K |
| Input $ / M tokens | — | $15 |
| Output $ / M tokens | — | $75 |
| Results tracked | 46 | 56 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Claude Opus 4 leads
Claude 3 Opus: 32.9 (#267), Claude Opus 4: 47.2 (#62)
| Benchmark | Claude 3 Opus | Claude Opus 4 |
|---|---|---|
| WeirdML | 19.2% | 43.7% |
| LMArena Coding | 1264 | 1442 |
| SWE-bench Verified | — | 70.7% |
| SWE-bench Verified (bash only) | — | 67.6% |
| Aider Polyglot | — | 72% |
| GSO | — | 6.9% |
| BigCodeBench Instruct | 45.5% | — |
| LiveBench Coding | 38.6% | — |
| BigCodeBench Complete | 57.4% | — |
| AlgoTune | — | 1.33 |
| HumanEval+ | 77.4% | — |
| MBPP+ | 73.3% | — |
Agentic & Tool Use Claude Opus 4 leads
Claude 3 Opus: 24.6 (#116), Claude Opus 4: 34.8 (#42)
| Benchmark | Claude 3 Opus | Claude Opus 4 |
|---|---|---|
| Cybench | 10% | 38% |
| METR Time Horizons | 29.5% | 63.9% |
| DeepResearch Bench | — | 46.8% |
| LMArena Search | — | 1127 |
Reasoning Claude Opus 4 leads
Claude 3 Opus: 14.6 (#324), Claude Opus 4: 27.3 (#121)
| Benchmark | Claude 3 Opus | Claude Opus 4 |
|---|---|---|
| SimpleBench | 23.5% | 58.8% |
| EnigmaEval | 0.8% | 5.6% |
| LMArena Hard Prompts | 1245 | 1399 |
| DTBench | 61.6% | 81.6% |
| LMCA | 17% | 37.4% |
| Epoch Capabilities Index | 126.91 | 142.67 |
| ForecastBench | 58.4 | 61.1 |
| ARC-AGI-2 | — | 8.6% |
| Kagi LLM Benchmark | — | 74.3% |
| ARC-AGI-1 | — | 35.7% |
| CritPt | — | 0.3% |
| Chess Puzzles | 5% | — |
| LiveBench Reasoning | 40.6% | — |
| LiveBench Data Analysis | 57.9% | — |
| LiveBench | 49.2% | — |
| WinoGrande | 88.5% | — |
Math Claude Opus 4 leads
Claude 3 Opus: 14.8 (#299), Claude Opus 4: 42.0 (#86)
| Benchmark | Claude 3 Opus | Claude Opus 4 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 4.7% | 64.4% |
| LMArena Math | 1273 | 1390 |
| MATH Level 5 | 37.5% | 85% |
| Omni-MATH | — | 61.6% |
| LiveBench Math | 43.6% | — |
| FrontierMath (Feb 2025 set) | — | 4.5% |
| FrontierMath Tier 4 (v1) | — | 4.2% |
Knowledge Claude Opus 4 leads
Claude 3 Opus: 24.5 (#267), Claude Opus 4: 44.0 (#88)
| Benchmark | Claude 3 Opus | Claude Opus 4 |
|---|---|---|
| GPQA Diamond | 47.2% | 76.3% |
| Confabulations | 22.7% | 15.9% |
| LMArena Expert | 1223 | 1386 |
| Humanity's Last Exam | — | 10.7% |
| SimpleQA Verified | 12.6% | — |
| MMLU-Pro | — | 87.5% |
| Vectara Hallucination Rate | — | 12% |
| GPQA (HELM) | — | 70.8% |
| MMLU | 84.6% | — |
Multimodal Claude Opus 4 leads
Claude 3 Opus: 27.1 (#116), Claude Opus 4: 31.5 (#106)
| Benchmark | Claude 3 Opus | Claude Opus 4 |
|---|---|---|
| LMArena Vision | 1023 | 1192 |
| GeoBench | — | 49% |
| VPCT | — | 38% |
Multilingual Claude Opus 4 leads
Claude 3 Opus: 41.4 (#207), Claude Opus 4: 48.8 (#138)
| Benchmark | Claude 3 Opus | Claude Opus 4 |
|---|---|---|
| LMArena Non-English | 1258 | 1362 |
| LMArena Chinese | 1248 | 1386 |
| LMArena French | 1275 | 1372 |
| LMArena German | 1258 | 1391 |
| LMArena Japanese | 1204 | 1331 |
| LMArena Korean | 1187 | 1321 |
| LMArena Russian | 1280 | 1392 |
| LMArena Spanish | 1246 | 1389 |
Instruction Following Claude Opus 4 leads
Claude 3 Opus: 64.1 (#228), Claude Opus 4: 77.1 (#28)
| Benchmark | Claude 3 Opus | Claude Opus 4 |
|---|---|---|
| LMArena Instruction Following | 1248 | 1406 |
| LiveBench Instruction Following | 63.9% | — |
| IFEval | — | 91.8% |
Long Context Claude Opus 4 leads
Claude 3 Opus: 38.2 (#202), Claude Opus 4: 39.6 (#172)
| Benchmark | Claude 3 Opus | Claude Opus 4 |
|---|---|---|
| LMArena Longer Query | 1259 | 1422 |
| Fiction.LiveBench | — | 61.1% |
Writing & Preference Claude Opus 4 leads
Claude 3 Opus: 47.2 (#213), Claude Opus 4: 61.2 (#89)
| Benchmark | Claude 3 Opus | Claude Opus 4 |
|---|---|---|
| LMArena Text | 1262 | 1377 |
| LMArena Creative Writing | 1235 | 1387 |
| LMArena Multi-Turn | 1275 | 1396 |
| Short-Story Creative Writing | — | 83.6% |
| EQ-Bench Creative Writing | — | 1580 |
| WildBench | — | 85.2% |
| LiveBench Language | 50.4% | — |
Frequently asked questions
Is Claude 3 Opus better than Claude Opus 4?
Claude Opus 4 is the stronger model overall, scoring 43.1 to 29.5 on the Noometry Index.
Is Claude 3 Opus or Claude Opus 4 better for coding?
Claude Opus 4 scores higher on coding benchmarks: 47.2 versus 32.9 in the Noometry coding category.
How many benchmarks do Claude 3 Opus and Claude Opus 4 share?
31 benchmarks have published results for both models. Claude 3 Opus has 46 scored results on Noometry and Claude Opus 4 has 56.