Model comparison
Claude 3 Opus vs o3
o3 is the stronger model overall, scoring 47.5 to 29.5 on the Noometry Index.
Last verified . 32 shared benchmarks.
Summary
- They share 32 benchmarks with published results for both. Claude 3 Opus scores higher in 0 categories and o3 in 10 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in math, where o3 leads 50.2 to 14.8.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 4.7% for Claude 3 Opus and 84.4% for o3.
Side by side
| Claude 3 Opus | o3 | |
|---|---|---|
| Provider | Anthropic | OpenAI |
| Noometry Index | 29.5 | 47.5 |
| Released | 2024-02-29 | 2025-04-16 |
| Weights | Proprietary | Proprietary |
| Context window | — | 200K |
| Max output | — | 100K |
| Input $ / M tokens | — | $2 |
| Output $ / M tokens | — | $8 |
| Results tracked | 46 | 63 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding o3 leads
Claude 3 Opus: 32.9 (#267), o3: 46.8 (#64)
| Benchmark | Claude 3 Opus | o3 |
|---|---|---|
| WeirdML | 19.2% | 52.4% |
| LMArena Coding | 1264 | 1408 |
| SWE-bench Verified | — | 62.3% |
| SWE-bench Verified (bash only) | — | 58.4% |
| Aider Polyglot | — | 81.3% |
| GSO | — | 8.8% |
| BigCodeBench Instruct | 45.5% | — |
| LiveBench Coding | 38.6% | — |
| BigCodeBench Complete | 57.4% | — |
| CadEval | — | 74% |
| ALE-Bench | — | 933.55 |
| HumanEval+ | 77.4% | — |
| MBPP+ | 73.3% | — |
Agentic & Tool Use o3 leads
Claude 3 Opus: 24.6 (#116), o3: 34.5 (#44)
| Benchmark | Claude 3 Opus | o3 |
|---|---|---|
| METR Time Horizons | 29.5% | 65.4% |
| Berkeley Function Calling Leaderboard | — | 63% |
| GDPval | — | 30.8% |
| Cybench | 10% | — |
| DeepResearch Bench | — | 45.2% |
| OSWorld | — | 23% |
| LMArena Search | — | 1144 |
Reasoning o3 leads
Claude 3 Opus: 14.6 (#324), o3: 32.0 (#78)
| Benchmark | Claude 3 Opus | o3 |
|---|---|---|
| SimpleBench | 23.5% | 53.1% |
| Chess Puzzles | 5% | 38% |
| EnigmaEval | 0.8% | 13.1% |
| LMArena Hard Prompts | 1245 | 1402 |
| DTBench | 61.6% | 84.8% |
| LMCA | 17% | 39.7% |
| Epoch Capabilities Index | 126.91 | 146.86 |
| ForecastBench | 58.4 | 62.5 |
| ARC-AGI-2 | — | 6.5% |
| Kagi LLM Benchmark | — | 67.6% |
| ARC-AGI-1 | — | 60.8% |
| CritPt | — | 1.4% |
| LiveBench Reasoning | 40.6% | — |
| Mystery Game Puzzles | — | 29% |
| LiveBench Data Analysis | 57.9% | — |
| LiveBench | 49.2% | — |
| WinoGrande | 88.5% | — |
Math o3 leads
Claude 3 Opus: 14.8 (#299), o3: 50.2 (#58)
| Benchmark | Claude 3 Opus | o3 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 4.7% | 84.4% |
| LMArena Math | 1273 | 1426 |
| MATH Level 5 | 37.5% | 97.8% |
| FrontierMath (Tiers 1-3) | — | 33.3% |
| Omni-MATH | — | 71.4% |
| LiveBench Math | 43.6% | — |
| FrontierMath (Feb 2025 set) | — | 18.7% |
| FrontierMath Tier 4 (v1) | — | 2.1% |
Knowledge o3 leads
Claude 3 Opus: 24.5 (#267), o3: 54.6 (#52)
| Benchmark | Claude 3 Opus | o3 |
|---|---|---|
| GPQA Diamond | 47.2% | 81.8% |
| SimpleQA Verified | 12.6% | 49.4% |
| Confabulations | 22.7% | 14.4% |
| LMArena Expert | 1223 | 1402 |
| Humanity's Last Exam | — | 20.3% |
| MMLU-Pro | — | 85.9% |
| GPQA (HELM) | — | 75.3% |
| MMLU | 84.6% | — |
Multimodal o3 leads
Claude 3 Opus: 27.1 (#116), o3: 41.4 (#36)
| Benchmark | Claude 3 Opus | o3 |
|---|---|---|
| LMArena Vision | 1023 | 1214 |
| GeoBench | — | 74% |
| VPCT | — | 52% |
Multilingual o3 leads
Claude 3 Opus: 41.4 (#207), o3: 51.7 (#105)
| Benchmark | Claude 3 Opus | o3 |
|---|---|---|
| LMArena Non-English | 1258 | 1401 |
| LMArena Chinese | 1248 | 1437 |
| LMArena French | 1275 | 1430 |
| LMArena German | 1258 | 1420 |
| LMArena Japanese | 1204 | 1403 |
| LMArena Korean | 1187 | 1370 |
| LMArena Russian | 1280 | 1406 |
| LMArena Spanish | 1246 | 1395 |
Instruction Following o3 leads
Claude 3 Opus: 64.1 (#228), o3: 72.8 (#127)
| Benchmark | Claude 3 Opus | o3 |
|---|---|---|
| LMArena Instruction Following | 1248 | 1368 |
| LiveBench Instruction Following | 63.9% | — |
| IFEval | — | 86.9% |
Long Context o3 leads
Claude 3 Opus: 38.2 (#202), o3: 53.3 (#6)
| Benchmark | Claude 3 Opus | o3 |
|---|---|---|
| LMArena Longer Query | 1259 | 1372 |
| Fiction.LiveBench | — | 88.9% |
| CL-bench | — | 17.8% |
Writing & Preference o3 leads
Claude 3 Opus: 47.2 (#213), o3: 63.5 (#64)
| Benchmark | Claude 3 Opus | o3 |
|---|---|---|
| LMArena Text | 1262 | 1410 |
| LMArena Creative Writing | 1235 | 1359 |
| LMArena Multi-Turn | 1275 | 1405 |
| Short-Story Creative Writing | — | 83.9% |
| EQ-Bench Creative Writing | — | 1676 |
| WildBench | — | 86.1% |
| LiveBench Language | 50.4% | — |
Frequently asked questions
Is Claude 3 Opus better than o3?
o3 is the stronger model overall, scoring 47.5 to 29.5 on the Noometry Index.
Is Claude 3 Opus or o3 better for coding?
o3 scores higher on coding benchmarks: 46.8 versus 32.9 in the Noometry coding category.
How many benchmarks do Claude 3 Opus and o3 share?
32 benchmarks have published results for both models. Claude 3 Opus has 46 scored results on Noometry and o3 has 63.