Model comparison
o1-mini vs o3
o3 is the stronger model overall, scoring 47.4 to 33.7 on the Noometry Index.
Last verified . 29 shared benchmarks.
Summary
- They share 29 benchmarks with published results for both. o1-mini scores higher in 0 categories and o3 in 9 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where o3 leads 32.1 to 8.9.
- The biggest single-benchmark swing is Aider Polyglot: 32.9% for o1-mini and 81.3% for o3.
Side by side
| o1-mini | o3 | |
|---|---|---|
| Provider | OpenAI | OpenAI |
| Noometry Index | 33.7 | 47.4 |
| Released | 2024-09-12 | 2025-04-16 |
| Weights | Proprietary | Proprietary |
| Context window | — | 200K |
| Max output | — | 100K |
| Input $ / M tokens | — | $2 |
| Output $ / M tokens | — | $8 |
| Results tracked | 39 | 63 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding o3 leads
o1-mini: 35.1 (#218), o3: 47.3 (#57)
| Benchmark | o1-mini | o3 |
|---|---|---|
| Aider Polyglot | 32.9% | 81.3% |
| WeirdML | 36.3% | 52.4% |
| LMArena Coding | 1362 | 1408 |
| SWE-bench Verified | — | 62.3% |
| SWE-bench Verified (bash only) | — | 58.4% |
| GSO | — | 8.8% |
| LiveBench Coding | 48% | — |
| CadEval | — | 74% |
| ALE-Bench | — | 933.55 |
| HumanEval+ | 89% | — |
| MBPP+ | 78.8% | — |
Agentic & Tool Use o3 leads
o1-mini: 24.3 (#112), o3: 34.3 (#42)
| Benchmark | o1-mini | o3 |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 63% |
| GDPval | — | 30.8% |
| Cybench | 10% | — |
| DeepResearch Bench | — | 45.2% |
| OSWorld | — | 23% |
| LMArena Search | — | 1144 |
| METR Time Horizons | — | 65.4% |
Reasoning o3 leads
o1-mini: 8.9 (#340), o3: 32.1 (#66)
| Benchmark | o1-mini | o3 |
|---|---|---|
| ARC-AGI-2 | 0.8% | 6.5% |
| SimpleBench | 18.1% | 53.1% |
| ARC-AGI-1 | 14% | 60.8% |
| LMArena Hard Prompts | 1333 | 1402 |
| Epoch Capabilities Index | 135.82 | 146.88 |
| Kagi LLM Benchmark | — | 67.6% |
| CritPt | — | 1.4% |
| Chess Puzzles | — | 38% |
| EnigmaEval | — | 13.1% |
| LiveBench Reasoning | 72.3% | — |
| Mystery Game Puzzles | — | 29% |
| DTBench | — | 84.8% |
| LiveBench Data Analysis | 57.9% | — |
| LMCA | — | 39.7% |
| ForecastBench | — | 62.5 |
| LiveBench | 57.8% | — |
Math o3 leads
o1-mini: 33.7 (#168), o3: 48.9 (#49)
| Benchmark | o1-mini | o3 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 46.9% | 84.4% |
| LMArena Math | 1358 | 1426 |
| MATH Level 5 | 89.2% | 97.8% |
| FrontierMath (Feb 2025 set) | 1.7% | 18.7% |
| FrontierMath (Tiers 1-3) | — | 33.3% |
| Omni-MATH | — | 71.4% |
| LiveBench Math | 62% | — |
| FrontierMath Tier 4 (v1) | — | 2.1% |
Knowledge o3 leads
o1-mini: 34.8 (#180), o3: 54.5 (#47)
| Benchmark | o1-mini | o3 |
|---|---|---|
| GPQA Diamond | 62.4% | 81.8% |
| Confabulations | 18.6% | 14.4% |
| LMArena Expert | 1316 | 1403 |
| Humanity's Last Exam | — | 20.3% |
| SimpleQA Verified | — | 49.4% |
| MMLU-Pro | — | 85.9% |
| GPQA (HELM) | — | 75.3% |
Multimodal Not comparable
o1-mini: —, o3: 40.7 (#35)
| Benchmark | o1-mini | o3 |
|---|---|---|
| LMArena Vision | — | 1214 |
| GeoBench | — | 74% |
| VPCT | — | 52% |
Multilingual o3 leads
o1-mini: 43.8 (#177), o3: 51.9 (#99)
| Benchmark | o1-mini | o3 |
|---|---|---|
| LMArena Non-English | 1289 | 1402 |
| LMArena Chinese | 1314 | 1438 |
| LMArena French | 1292 | 1430 |
| LMArena German | 1278 | 1420 |
| LMArena Japanese | 1245 | 1404 |
| LMArena Korean | 1223 | 1371 |
| LMArena Russian | 1283 | 1407 |
| LMArena Spanish | 1302 | 1395 |
Instruction Following o3 leads
o1-mini: 67.0 (#200), o3: 73.0 (#122)
| Benchmark | o1-mini | o3 |
|---|---|---|
| LMArena Instruction Following | 1304 | 1368 |
| LiveBench Instruction Following | 65.4% | — |
| IFEval | — | 86.9% |
Long Context o3 leads
o1-mini: 40.1 (#156), o3: 53.2 (#6)
| Benchmark | o1-mini | o3 |
|---|---|---|
| LMArena Longer Query | 1321 | 1372 |
| Fiction.LiveBench | — | 88.9% |
| CL-bench | — | 17.8% |
Writing & Preference o3 leads
o1-mini: 48.6 (#196), o3: 63.7 (#62)
| Benchmark | o1-mini | o3 |
|---|---|---|
| LMArena Text | 1317 | 1410 |
| LMArena Creative Writing | 1244 | 1359 |
| Short-Story Creative Writing | 64.9% | 83.9% |
| LMArena Multi-Turn | 1314 | 1405 |
| EQ-Bench Creative Writing | — | 1676 |
| WildBench | — | 86.1% |
| LiveBench Language | 40.9% | — |
Frequently asked questions
Is o1-mini better than o3?
o3 is the stronger model overall, scoring 47.4 to 33.7 on the Noometry Index.
Is o1-mini or o3 better for coding?
o3 scores higher on coding benchmarks: 47.3 versus 35.1 in the Noometry coding category.
How many benchmarks do o1-mini and o3 share?
29 benchmarks have published results for both models. o1-mini has 39 scored results on Noometry and o3 has 63.