Model comparison
Claude 3.5 Sonnet vs o3
o3 is the stronger model overall, scoring 47.4 to 34.4 on the Noometry Index.
Last verified . 44 shared benchmarks.
Summary
- They share 44 benchmarks with published results for both. Claude 3.5 Sonnet 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 48.9 to 18.0.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 8.5% for Claude 3.5 Sonnet and 84.4% for o3.
Side by side
| Claude 3.5 Sonnet | o3 | |
|---|---|---|
| Provider | Anthropic | OpenAI |
| Noometry Index | 34.4 | 47.4 |
| Released | 2024-06-20 | 2025-04-16 |
| Weights | Proprietary | Proprietary |
| Context window | — | 200K |
| Max output | — | 100K |
| Input $ / M tokens | — | $2 |
| Output $ / M tokens | — | $8 |
| Results tracked | 60 | 63 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding o3 leads
Claude 3.5 Sonnet: 39.5 (#145), o3: 47.3 (#57)
| Benchmark | Claude 3.5 Sonnet | o3 |
|---|---|---|
| Aider Polyglot | 51.6% | 81.3% |
| GSO | 4.6% | 8.8% |
| WeirdML | 40% | 52.4% |
| LMArena Coding | 1343 | 1408 |
| CadEval | 48% | 74% |
| SWE-bench Verified | — | 62.3% |
| SWE-bench Verified (bash only) | — | 58.4% |
| BigCodeBench Instruct | 46.8% | — |
| LiveBench Coding | 67.1% | — |
| BigCodeBench Complete | 58.6% | — |
| ALE-Bench | — | 933.55 |
| HumanEval+ | 81.7% | — |
| MBPP+ | 74.3% | — |
Agentic & Tool Use o3 leads
Claude 3.5 Sonnet: 32.1 (#61), o3: 34.3 (#42)
| Benchmark | Claude 3.5 Sonnet | o3 |
|---|---|---|
| METR Time Horizons | 45.2% | 65.4% |
| Berkeley Function Calling Leaderboard | — | 63% |
| GDPval | — | 30.8% |
| TheAgentCompany | 24% | — |
| Cybench | 17.5% | — |
| DeepResearch Bench | — | 45.2% |
| OSWorld | — | 23% |
| BALROG | 32.6% | — |
| LMArena Search | — | 1144 |
Reasoning o3 leads
Claude 3.5 Sonnet: 22.5 (#176), o3: 32.1 (#66)
| Benchmark | Claude 3.5 Sonnet | o3 |
|---|---|---|
| SimpleBench | 41.4% | 53.1% |
| EnigmaEval | 0.9% | 13.1% |
| LMArena Hard Prompts | 1305 | 1402 |
| DTBench | 67.8% | 84.8% |
| Epoch Capabilities Index | 133.55 | 146.88 |
| ForecastBench | 60.7 | 62.5 |
| ARC-AGI-2 | — | 6.5% |
| Kagi LLM Benchmark | — | 67.6% |
| ARC-AGI-1 | — | 60.8% |
| CritPt | — | 1.4% |
| Chess Puzzles | — | 38% |
| LiveBench Reasoning | 56.7% | — |
| Mystery Game Puzzles | — | 29% |
| LiveBench Data Analysis | 55% | — |
| LMCA | — | 39.7% |
| LiveBench | 59% | — |
Math o3 leads
Claude 3.5 Sonnet: 18.0 (#282), o3: 48.9 (#49)
| Benchmark | Claude 3.5 Sonnet | o3 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 8.5% | 84.4% |
| Omni-MATH | 27.6% | 71.4% |
| LMArena Math | 1306 | 1426 |
| MATH Level 5 | 56.9% | 97.8% |
| FrontierMath (Feb 2025 set) | 2.1% | 18.7% |
| FrontierMath Tier 4 (v1) | 0% | 2.1% |
| FrontierMath (Tiers 1-3) | — | 33.3% |
| LiveBench Math | 52.3% | — |
Knowledge o3 leads
Claude 3.5 Sonnet: 28.5 (#236), o3: 54.5 (#47)
| Benchmark | Claude 3.5 Sonnet | o3 |
|---|---|---|
| GPQA Diamond | 55.3% | 81.8% |
| Humanity's Last Exam | 4.1% | 20.3% |
| MMLU-Pro | 77.7% | 85.9% |
| Confabulations | 19.9% | 14.4% |
| GPQA (HELM) | 56.5% | 75.3% |
| LMArena Expert | 1264 | 1403 |
| SimpleQA Verified | — | 49.4% |
| MMLU | 87.3% | — |
Multimodal o3 leads
Claude 3.5 Sonnet: 26.0 (#115), o3: 40.7 (#35)
| Benchmark | Claude 3.5 Sonnet | o3 |
|---|---|---|
| LMArena Vision | 1125 | 1214 |
| GeoBench | 62% | 74% |
| VPCT | 33% | 52% |
| Video-MME | 60% | — |
Multilingual o3 leads
Claude 3.5 Sonnet: 43.4 (#180), o3: 51.9 (#99)
| Benchmark | Claude 3.5 Sonnet | o3 |
|---|---|---|
| LMArena Non-English | 1283 | 1402 |
| LMArena Chinese | 1271 | 1438 |
| LMArena French | 1304 | 1430 |
| LMArena German | 1297 | 1420 |
| LMArena Japanese | 1234 | 1404 |
| LMArena Korean | 1200 | 1371 |
| LMArena Russian | 1305 | 1407 |
| LMArena Spanish | 1289 | 1395 |
Instruction Following o3 leads
Claude 3.5 Sonnet: 69.0 (#177), o3: 73.0 (#122)
| Benchmark | Claude 3.5 Sonnet | o3 |
|---|---|---|
| IFEval | 85.5% | 86.9% |
| LMArena Instruction Following | 1297 | 1368 |
| LiveBench Instruction Following | 69.3% | — |
Long Context o3 leads
Claude 3.5 Sonnet: 39.9 (#162), o3: 53.2 (#6)
| Benchmark | Claude 3.5 Sonnet | o3 |
|---|---|---|
| LMArena Longer Query | 1311 | 1372 |
| Fiction.LiveBench | — | 88.9% |
| CL-bench | — | 17.8% |
Writing & Preference o3 leads
Claude 3.5 Sonnet: 53.0 (#159), o3: 63.7 (#62)
| Benchmark | Claude 3.5 Sonnet | o3 |
|---|---|---|
| LMArena Text | 1298 | 1410 |
| LMArena Creative Writing | 1292 | 1359 |
| Short-Story Creative Writing | 80.3% | 83.9% |
| EQ-Bench Creative Writing | 1451 | 1676 |
| WildBench | 79.2% | 86.1% |
| LMArena Multi-Turn | 1326 | 1405 |
| LiveBench Language | 53.8% | — |
Frequently asked questions
Is Claude 3.5 Sonnet better than o3?
o3 is the stronger model overall, scoring 47.4 to 34.4 on the Noometry Index.
Is Claude 3.5 Sonnet or o3 better for coding?
o3 scores higher on coding benchmarks: 47.3 versus 39.5 in the Noometry coding category.
How many benchmarks do Claude 3.5 Sonnet and o3 share?
44 benchmarks have published results for both models. Claude 3.5 Sonnet has 60 scored results on Noometry and o3 has 63.