Model comparison
MiniMax-M3 vs o1-mini
MiniMax-M3 is the stronger model overall, scoring 43.5 to 33.7 on the Noometry Index.
Last verified . 21 shared benchmarks.
MiniMaxSummary
- They share 21 benchmarks with published results for both. MiniMax-M3 scores higher in 8 categories and o1-mini in 1 category; 9 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where MiniMax-M3 leads 58.4 to 34.8.
- The biggest single-benchmark swing is GPQA Diamond: 90.9% for MiniMax-M3 and 62.4% for o1-mini.
- MiniMax-M3 has downloadable open weights; the other is API-only.
Side by side
| MiniMax-M3 | o1-mini | |
|---|---|---|
| Provider | MiniMax | OpenAI |
| Noometry Index | 43.5 | 33.7 |
| Released | 2026-06-01 | 2024-09-12 |
| Weights | Open | Proprietary |
| Context window | 1M | — |
| Max output | 512K | — |
| Input $ / M tokens | $0.30 | — |
| Output $ / M tokens | $1.20 | — |
| Results tracked | 41 | 39 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding MiniMax-M3 leads
MiniMax-M3: 41.5 (#111), o1-mini: 35.1 (#218)
| Benchmark | MiniMax-M3 | o1-mini |
|---|---|---|
| LMArena Coding | 1470 | 1362 |
| FrontierCode | 14.7% | — |
| Aider Polyglot | — | 32.9% |
| LMArena WebDev | 1482 | — |
| SciCode | 47.1% | — |
| WeirdML | — | 36.3% |
| LiveBench Coding | — | 48% |
| ALE-Bench | 640.02 | — |
| HumanEval+ | — | 89% |
| MBPP+ | — | 78.8% |
Agentic & Tool Use o1-mini leads
MiniMax-M3: 22.5 (#121), o1-mini: 24.3 (#112)
| Benchmark | MiniMax-M3 | o1-mini |
|---|---|---|
| APEX-Agents | 37.7% | — |
| OSWorld 2.0 | 4.6% | — |
| Cybench | — | 10% |
| GBAEval | 0.9% | — |
| Vending-Bench 2 | 2,158 | — |
Reasoning MiniMax-M3 leads
MiniMax-M3: 29.3 (#83), o1-mini: 8.9 (#340)
| Benchmark | MiniMax-M3 | o1-mini |
|---|---|---|
| SimpleBench | 45.8% | 18.1% |
| LMArena Hard Prompts | 1447 | 1333 |
| Epoch Capabilities Index | 147 | 135.82 |
| ARC-AGI-2 | — | 0.8% |
| NYT Connections (extended) | 65.1% | — |
| ARC-AGI-1 | — | 14% |
| CritPt | 3.7% | — |
| Chess Puzzles | 14% | — |
| LiveBench Reasoning | — | 72.3% |
| Mystery Game Puzzles | 8% | — |
| DTBench | 78.9% | — |
| LiveBench Data Analysis | — | 57.9% |
| LMCA | 33.7% | — |
| Surface Evolver Bench | 55% | — |
| ForecastBench | 61.4 | — |
| LiveBench | — | 57.8% |
Math MiniMax-M3 leads
MiniMax-M3: 39.1 (#83), o1-mini: 33.7 (#168)
| Benchmark | MiniMax-M3 | o1-mini |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 71.1% | 46.9% |
| LMArena Math | 1432 | 1358 |
| ProofBench | 18% | — |
| LiveBench Math | — | 62% |
| MATH Level 5 | — | 89.2% |
| FrontierMath (Feb 2025 set) | — | 1.7% |
Knowledge MiniMax-M3 leads
MiniMax-M3: 58.4 (#32), o1-mini: 34.8 (#180)
| Benchmark | MiniMax-M3 | o1-mini |
|---|---|---|
| GPQA Diamond | 90.9% | 62.4% |
| LMArena Expert | 1462 | 1316 |
| Confabulations | — | 18.6% |
Multimodal Not comparable
MiniMax-M3: 39.5 (#45), o1-mini: —
| Benchmark | MiniMax-M3 | o1-mini |
|---|---|---|
| LMArena Vision | 1254 | — |
| LMArena Document | 1435 | — |
Multilingual MiniMax-M3 leads
MiniMax-M3: 53.2 (#70), o1-mini: 43.8 (#177)
| Benchmark | MiniMax-M3 | o1-mini |
|---|---|---|
| LMArena Non-English | 1420 | 1289 |
| LMArena Chinese | 1465 | 1314 |
| LMArena French | 1445 | 1292 |
| LMArena German | 1424 | 1278 |
| LMArena Japanese | 1381 | 1245 |
| LMArena Korean | 1372 | 1223 |
| LMArena Russian | 1430 | 1283 |
| LMArena Spanish | 1431 | 1302 |
Instruction Following MiniMax-M3 leads
MiniMax-M3: 75.7 (#56), o1-mini: 67.0 (#200)
| Benchmark | MiniMax-M3 | o1-mini |
|---|---|---|
| LMArena Instruction Following | 1433 | 1304 |
| LiveBench Instruction Following | — | 65.4% |
Long Context MiniMax-M3 leads
MiniMax-M3: 44.2 (#68), o1-mini: 40.1 (#156)
| Benchmark | MiniMax-M3 | o1-mini |
|---|---|---|
| LMArena Longer Query | 1445 | 1321 |
Writing & Preference MiniMax-M3 leads
MiniMax-M3: 62.3 (#79), o1-mini: 48.6 (#196)
| Benchmark | MiniMax-M3 | o1-mini |
|---|---|---|
| LMArena Text | 1433 | 1317 |
| LMArena Creative Writing | 1405 | 1244 |
| LMArena Multi-Turn | 1441 | 1314 |
| Short-Story Creative Writing | — | 64.9% |
| EQ-Bench 4 | 1150 | — |
| LiveBench Language | — | 40.9% |
Frequently asked questions
Is MiniMax-M3 better than o1-mini?
MiniMax-M3 is the stronger model overall, scoring 43.5 to 33.7 on the Noometry Index.
Is MiniMax-M3 or o1-mini better for coding?
MiniMax-M3 scores higher on coding benchmarks: 41.5 versus 35.1 in the Noometry coding category.
How many benchmarks do MiniMax-M3 and o1-mini share?
21 benchmarks have published results for both models. MiniMax-M3 has 41 scored results on Noometry and o1-mini has 39.