Model comparison
o1 vs o4-mini
o4-mini is the stronger model overall, scoring 41.6 to 40.6 on the Noometry Index.
Last verified . 41 shared benchmarks.
Summary
- They share 41 benchmarks with published results for both. o1 scores higher in 5 categories and o4-mini in 5 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in agentic & tool use, where o4-mini leads 32.3 to 24.3.
- The biggest single-benchmark swing is ARC-AGI-1: 30.7% for o1 and 58.7% for o4-mini.
- o4-mini is cheaper at $1.10 / $4.40 per million input/output tokens, against $15 / $60 for o1.
Side by side
| o1 | o4-mini | |
|---|---|---|
| Provider | OpenAI | OpenAI |
| Noometry Index | 40.6 | 41.6 |
| Released | 2024-09-12 | 2025-04-16 |
| Weights | Proprietary | Proprietary |
| Context window | 200K | 200K |
| Max output | 100K | 100K |
| Input $ / M tokens | $15 | $1.10 |
| Output $ / M tokens | $60 | $4.40 |
| Results tracked | 52 | 60 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding o1 leads
o1: 45.7 (#64), o4-mini: 41.6 (#110)
| Benchmark | o1 | o4-mini |
|---|---|---|
| Aider Polyglot | 61.7% | 72% |
| WeirdML | 47.6% | 52.6% |
| LMArena Coding | 1367 | 1369 |
| CadEval | 56% | 62% |
| SWE-bench Verified (bash only) | — | 45% |
| GSO | — | 3.6% |
| LiveBench Coding | 69.7% | — |
| ALE-Bench | — | 826.17 |
| AlgoTune | — | 1.72 |
| HumanEval+ | 89% | — |
| MBPP+ | 80.2% | — |
Agentic & Tool Use o4-mini leads
o1: 24.3 (#111), o4-mini: 32.3 (#57)
| Benchmark | o1 | o4-mini |
|---|---|---|
| METR Time Horizons | 51.1% | 63.9% |
| Berkeley Function Calling Leaderboard | — | 53.2% |
| GDPval | — | 25.3% |
| Cybench | 10% | — |
Reasoning o1 leads
o1: 27.5 (#103), o4-mini: 24.6 (#143)
| Benchmark | o1 | o4-mini |
|---|---|---|
| SimpleBench | 41.7% | 38.7% |
| ARC-AGI-1 | 30.7% | 58.7% |
| Chess Puzzles | 15% | 26% |
| EnigmaEval | 5.7% | 9.2% |
| LMArena Hard Prompts | 1372 | 1351 |
| DTBench | 74.7% | 77.6% |
| LMCA | 22.3% | 26.5% |
| Epoch Capabilities Index | 141.92 | 145.63 |
| ARC-AGI-2 | — | 6.1% |
| Kagi LLM Benchmark | — | 67.6% |
| CritPt | — | 0.6% |
| LiveBench Reasoning | 91.6% | — |
| Mystery Game Puzzles | — | 5% |
| LiveBench Data Analysis | 65.5% | — |
| ForecastBench | — | 61.8 |
| LiveBench | 75.7% | — |
Math o4-mini leads
o1: 34.9 (#150), o4-mini: 40.2 (#79)
| Benchmark | o1 | o4-mini |
|---|---|---|
| FrontierMath (Tiers 1-3) | 14.7% | 36.1% |
| OTIS Mock AIME 2024-2025 | 73.3% | 81.7% |
| LMArena Math | 1388 | 1390 |
| MATH Level 5 | 94.7% | 97.8% |
| FrontierMath (Feb 2025 set) | 9.3% | 24.8% |
| FrontierMath Tier 4 | — | 4.9% |
| Omni-MATH | — | 72% |
| LiveBench Math | 80.3% | — |
| FrontierMath Tier 4 (v1) | — | 6.3% |
Knowledge o4-mini leads
o1: 41.5 (#101), o4-mini: 43.5 (#83)
| Benchmark | o1 | o4-mini |
|---|---|---|
| GPQA Diamond | 76.8% | 79.6% |
| Humanity's Last Exam | 8% | 18.1% |
| SimpleQA Verified | 41.1% | 19.6% |
| Confabulations | 11.7% | 15.8% |
| LMArena Expert | 1362 | 1344 |
| MMLU-Pro | — | 82% |
| Vectara Hallucination Rate | — | 18.6% |
| GPQA (HELM) | — | 73.5% |
Multimodal o4-mini leads
o1: 33.6 (#87), o4-mini: 39.5 (#46)
| Benchmark | o1 | o4-mini |
|---|---|---|
| LMArena Vision | 1168 | 1194 |
| GeoBench | 80% | 64% |
| VPCT | 37% | 57.5% |
| SpatialViz-Bench | 41.4% | — |
Multilingual o1 leads
o1: 48.8 (#137), o4-mini: 47.2 (#149)
| Benchmark | o1 | o4-mini |
|---|---|---|
| LMArena Non-English | 1359 | 1336 |
| LMArena Chinese | 1394 | 1355 |
| LMArena French | 1343 | 1362 |
| LMArena German | 1337 | 1336 |
| LMArena Japanese | 1346 | 1308 |
| LMArena Korean | 1396 | 1313 |
| LMArena Russian | 1356 | 1334 |
| LMArena Spanish | 1345 | 1346 |
Instruction Following Too close to call
o1: 75.0 (#82), o4-mini: 75.4 (#63)
| Benchmark | o1 | o4-mini |
|---|---|---|
| LMArena Instruction Following | 1368 | 1321 |
| LiveBench Instruction Following | 81.5% | — |
| IFEval | — | 92.8% |
Long Context o1 leads
o1: 50.3 (#9), o4-mini: 45.5 (#31)
| Benchmark | o1 | o4-mini |
|---|---|---|
| Fiction.LiveBench | 83.3% | 77.8% |
| LMArena Longer Query | 1378 | 1315 |
Writing & Preference o1 leads
o1: 55.8 (#139), o4-mini: 54.2 (#148)
| Benchmark | o1 | o4-mini |
|---|---|---|
| LMArena Text | 1366 | 1353 |
| LMArena Creative Writing | 1347 | 1294 |
| Short-Story Creative Writing | 70.2% | 75% |
| LMArena Multi-Turn | 1368 | 1350 |
| WildBench | — | 85.4% |
| LiveBench Language | 65.4% | — |
Frequently asked questions
Is o1 better than o4-mini?
o4-mini is the stronger model overall, scoring 41.6 to 40.6 on the Noometry Index.
Which is cheaper, o1 or o4-mini?
o4-mini is cheaper. It lists at $1.10 per million input tokens and $4.40 per million output tokens; o1 lists at $15 and $60.
Is o1 or o4-mini better for coding?
o1 scores higher on coding benchmarks: 45.7 versus 41.6 in the Noometry coding category.
Which has the bigger context window?
Both accept 200K tokens.
How many benchmarks do o1 and o4-mini share?
41 benchmarks have published results for both models. o1 has 52 scored results on Noometry and o4-mini has 60.