Model comparison
o1-mini vs o3-mini
o3-mini is the stronger model overall, scoring 36.7 to 34.0 on the Noometry Index.
Last verified . 37 shared benchmarks.
Summary
- They share 37 benchmarks with published results for both. o1-mini scores higher in 2 categories and o3-mini in 7 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in instruction following, where o3-mini leads 75.1 to 66.7.
- The biggest single-benchmark swing is LiveBench Coding: 48% for o1-mini and 82.7% for o3-mini.
Side by side
| o1-mini | o3-mini | |
|---|---|---|
| Provider | OpenAI | OpenAI |
| Noometry Index | 34.0 | 36.7 |
| Released | 2024-09-12 | 2024-12-20 |
| Weights | Proprietary | Proprietary |
| Context window | — | 200K |
| Max output | — | 100K |
| Input $ / M tokens | — | $1.10 |
| Output $ / M tokens | — | $4.40 |
| Results tracked | 39 | 51 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding o3-mini leads
o1-mini: 35.5 (#224), o3-mini: 40.8 (#132)
| Benchmark | o1-mini | o3-mini |
|---|---|---|
| Aider Polyglot | 32.9% | 60.4% |
| WeirdML | 36.3% | 43.7% |
| LiveBench Coding | 48% | 82.7% |
| LMArena Coding | 1362 | 1378 |
| SciCode | — | 39.8% |
| GSO | — | 1.3% |
| CadEval | — | 54% |
| HumanEval+ | 89% | — |
| MBPP+ | 78.8% | — |
Agentic & Tool Use o3-mini leads
o1-mini: 24.6 (#118), o3-mini: 29.6 (#84)
| Benchmark | o1-mini | o3-mini |
|---|---|---|
| Cybench | 10% | 22.5% |
Reasoning o3-mini leads
o1-mini: 8.8 (#346), o3-mini: 16.3 (#305)
| Benchmark | o1-mini | o3-mini |
|---|---|---|
| ARC-AGI-2 | 0.8% | 3% |
| SimpleBench | 18.1% | 22.8% |
| ARC-AGI-1 | 14% | 34.5% |
| LiveBench Reasoning | 72.3% | 89.6% |
| LMArena Hard Prompts | 1333 | 1366 |
| LiveBench Data Analysis | 57.9% | 70.6% |
| Epoch Capabilities Index | 135.82 | 140.34 |
| LiveBench | 57.8% | 75.9% |
| CritPt | — | 0.3% |
| Chess Puzzles | — | 17% |
| Mystery Game Puzzles | — | 7% |
| DTBench | — | 68.8% |
| LMCA | — | 19% |
| ForecastBench | — | 59.6 |
Math o1-mini leads
o1-mini: 35.4 (#186), o3-mini: 28.1 (#244)
| Benchmark | o1-mini | o3-mini |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 46.9% | 76.9% |
| LiveBench Math | 62% | 77.3% |
| LMArena Math | 1358 | 1396 |
| MATH Level 5 | 89.2% | 96.5% |
| FrontierMath (Feb 2025 set) | 1.7% | 12.4% |
| FrontierMath (Tiers 1-3) | — | 18.6% |
| FrontierMath Tier 4 | — | 0% |
| FrontierMath Tier 4 (v1) | — | 4.2% |
Knowledge o3-mini leads
o1-mini: 34.9 (#192), o3-mini: 38.3 (#146)
| Benchmark | o1-mini | o3-mini |
|---|---|---|
| GPQA Diamond | 62.4% | 77% |
| Confabulations | 18.6% | 17.9% |
| LMArena Expert | 1316 | 1364 |
| SimpleQA Verified | — | 15.3% |
Multilingual o3-mini leads
o1-mini: 43.6 (#182), o3-mini: 45.7 (#164)
| Benchmark | o1-mini | o3-mini |
|---|---|---|
| LMArena Non-English | 1289 | 1319 |
| LMArena Chinese | 1314 | 1379 |
| LMArena French | 1293 | 1334 |
| LMArena German | 1278 | 1303 |
| LMArena Japanese | 1245 | 1286 |
| LMArena Korean | 1223 | 1314 |
| LMArena Russian | 1283 | 1304 |
| LMArena Spanish | 1303 | 1321 |
Instruction Following o3-mini leads
o1-mini: 66.7 (#206), o3-mini: 75.1 (#72)
| Benchmark | o1-mini | o3-mini |
|---|---|---|
| LiveBench Instruction Following | 65.4% | 84.4% |
| LMArena Instruction Following | 1304 | 1337 |
Long Context o1-mini leads
o1-mini: 40.1 (#161), o3-mini: 33.8 (#256)
| Benchmark | o1-mini | o3-mini |
|---|---|---|
| LMArena Longer Query | 1320 | 1343 |
| Fiction.LiveBench | — | 50% |
Writing & Preference o3-mini leads
o1-mini: 48.4 (#202), o3-mini: 50.3 (#182)
| Benchmark | o1-mini | o3-mini |
|---|---|---|
| LMArena Text | 1317 | 1337 |
| LMArena Creative Writing | 1244 | 1286 |
| Short-Story Creative Writing | 64.9% | 61.7% |
| LMArena Multi-Turn | 1314 | 1320 |
| LiveBench Language | 40.9% | 50.7% |
Frequently asked questions
Is o1-mini better than o3-mini?
o3-mini is the stronger model overall, scoring 36.7 to 34.0 on the Noometry Index.
Is o1-mini or o3-mini better for coding?
o3-mini scores higher on coding benchmarks: 40.8 versus 35.5 in the Noometry coding category.
How many benchmarks do o1-mini and o3-mini share?
37 benchmarks have published results for both models. o1-mini has 39 scored results on Noometry and o3-mini has 51.