Model comparison
Claude 3 Opus vs Claude Opus 5
Claude Opus 5 is the stronger model overall, scoring 67.8 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 0 categories and Claude Opus 5 in 10 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in math, where Claude Opus 5 leads 86.2 to 14.8.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 4.7% for Claude 3 Opus and 98.9% for Claude Opus 5.
Side by side
| Claude 3 Opus | Claude Opus 5 | |
|---|---|---|
| Provider | Anthropic | Anthropic |
| Noometry Index | 29.5 | 67.8 |
| Released | 2024-02-29 | 2026-07-24 |
| Weights | Proprietary | Proprietary |
| Context window | — | 1M |
| Max output | — | 128K |
| Input $ / M tokens | — | $5 |
| Output $ / M tokens | — | $25 |
| Results tracked | 46 | 57 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Claude Opus 5 leads
Claude 3 Opus: 32.9 (#267), Claude Opus 5: 67.5 (#5)
| Benchmark | Claude 3 Opus | Claude Opus 5 |
|---|---|---|
| WeirdML | 19.2% | 91.8% |
| LMArena Coding | 1264 | 1534 |
| DeepSWE | — | 73.6% |
| FrontierCode | — | 53.4% |
| CursorBench | — | 46.6% |
| LMArena WebDev | — | 1691 |
| FrontierSWE | — | 52% |
| SciCode | — | 56.4% |
| BigCodeBench Instruct | 45.5% | — |
| LiveBench Coding | 38.6% | — |
| BigCodeBench Complete | 57.4% | — |
| ALE-Bench | — | 2,165 |
| HumanEval+ | 77.4% | — |
| MBPP+ | 73.3% | — |
Agentic & Tool Use Claude Opus 5 leads
Claude 3 Opus: 24.6 (#116), Claude Opus 5: 55.6 (#1)
| Benchmark | Claude 3 Opus | Claude Opus 5 |
|---|---|---|
| APEX-Agents | — | 65.8% |
| OSWorld 2.0 | — | 31.4% |
| τ²-bench Banking | — | 48.7% |
| Cybench | 10% | — |
| PostTrainBench | — | 35% |
| BALROG | — | 63.4% |
| GBAEval | — | 79.6% |
| GDP.pdf | — | 24% |
| METR Time Horizons | 29.5% | — |
| Vending-Bench 2 | — | 11,182 |
Reasoning Claude Opus 5 leads
Claude 3 Opus: 14.6 (#324), Claude Opus 5: 77.2 (#4)
| Benchmark | Claude 3 Opus | Claude Opus 5 |
|---|---|---|
| SimpleBench | 23.5% | 80.6% |
| Chess Puzzles | 5% | 42% |
| LMArena Hard Prompts | 1245 | 1526 |
| DTBench | 61.6% | 97.9% |
| LMCA | 17% | 64.5% |
| Epoch Capabilities Index | 126.91 | 162.78 |
| ARC-AGI-2 | — | 90.4% |
| NYT Connections (extended) | — | 94.3% |
| ARC-AGI-1 | — | 97.5% |
| CritPt | — | 29.1% |
| EnigmaEval | 0.8% | — |
| EBR-Bench | — | 45.7% |
| LiveBench Reasoning | 40.6% | — |
| Mystery Game Puzzles | — | 59% |
| LiveBench Data Analysis | 57.9% | — |
| Bench to the Future 3 | — | 0.12 |
| ForecastBench | 58.4 | — |
| LiveBench | 49.2% | — |
| WinoGrande | 88.5% | — |
Math Claude Opus 5 leads
Claude 3 Opus: 14.8 (#299), Claude Opus 5: 86.2 (#8)
| Benchmark | Claude 3 Opus | Claude Opus 5 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 4.7% | 98.9% |
| LMArena Math | 1273 | 1531 |
| FrontierMath (Tiers 1-3) | — | 85.6% |
| FrontierMath Tier 4 | — | 73.2% |
| ProofBench | — | 99% |
| LiveBench Math | 43.6% | — |
| MATH Level 5 | 37.5% | — |
Knowledge Claude Opus 5 leads
Claude 3 Opus: 24.5 (#267), Claude Opus 5: 66.8 (#9)
| Benchmark | Claude 3 Opus | Claude Opus 5 |
|---|---|---|
| GPQA Diamond | 47.2% | 93.9% |
| SimpleQA Verified | 12.6% | 59.9% |
| LMArena Expert | 1223 | 1557 |
| Confabulations | 22.7% | — |
| MMLU | 84.6% | — |
Multimodal Claude Opus 5 leads
Claude 3 Opus: 27.1 (#116), Claude Opus 5: 50.8 (#8)
| Benchmark | Claude 3 Opus | Claude Opus 5 |
|---|---|---|
| LMArena Vision | 1023 | 1319 |
| Blueprint-Bench 2 | — | 30.4% |
| Furniture Assembly | — | 60.8% |
| LMArena Document | — | 1516 |
Multilingual Claude Opus 5 leads
Claude 3 Opus: 41.4 (#207), Claude Opus 5: 58.8 (#4)
| Benchmark | Claude 3 Opus | Claude Opus 5 |
|---|---|---|
| LMArena Non-English | 1258 | 1501 |
| LMArena Chinese | 1248 | 1574 |
| LMArena French | 1275 | 1519 |
| LMArena German | 1258 | 1524 |
| LMArena Japanese | 1204 | 1516 |
| LMArena Korean | 1187 | 1521 |
| LMArena Russian | 1280 | 1507 |
| LMArena Spanish | 1246 | 1519 |
Instruction Following Claude Opus 5 leads
Claude 3 Opus: 64.1 (#228), Claude Opus 5: 79.2 (#7)
| Benchmark | Claude 3 Opus | Claude Opus 5 |
|---|---|---|
| LMArena Instruction Following | 1248 | 1517 |
| LiveBench Instruction Following | 63.9% | — |
Long Context Claude Opus 5 leads
Claude 3 Opus: 38.2 (#202), Claude Opus 5: 46.5 (#21)
| Benchmark | Claude 3 Opus | Claude Opus 5 |
|---|---|---|
| LMArena Longer Query | 1259 | 1515 |
Writing & Preference Claude Opus 5 leads
Claude 3 Opus: 47.2 (#213), Claude Opus 5: 79.2 (#1)
| Benchmark | Claude 3 Opus | Claude Opus 5 |
|---|---|---|
| LMArena Text | 1262 | 1507 |
| LMArena Creative Writing | 1235 | 1491 |
| LMArena Multi-Turn | 1275 | 1499 |
| EQ-Bench Creative Writing | — | 2133 |
| EQ-Bench 4 | — | 1385 |
| LiveBench Language | 50.4% | — |
Frequently asked questions
Is Claude 3 Opus better than Claude Opus 5?
Claude Opus 5 is the stronger model overall, scoring 67.8 to 29.5 on the Noometry Index.
Is Claude 3 Opus or Claude Opus 5 better for coding?
Claude Opus 5 scores higher on coding benchmarks: 67.5 versus 32.9 in the Noometry coding category.
How many benchmarks do Claude 3 Opus and Claude Opus 5 share?
27 benchmarks have published results for both models. Claude 3 Opus has 46 scored results on Noometry and Claude Opus 5 has 57.