Model comparison
MiniMax-M2.5 vs o3-pro
o3-pro is the stronger model overall, scoring 42.9 to 38.3 on the Noometry Index. MiniMax-M2.5 costs 67× less per token, which makes it the better buy when o3-pro's lead doesn't matter for your workload.
Last verified . 5 shared benchmarks.
MiniMaxSummary
- They share 5 benchmarks with published results for both. MiniMax-M2.5 scores higher in 1 category and o3-pro in 4 categories; 5 gaps are clear of the uncertainty.
- The widest gap is in long context, where o3-pro leads 72.2 to 37.5.
- The biggest single-benchmark swing is Kagi LLM Benchmark: 55.2% for MiniMax-M2.5 and 72.1% for o3-pro.
- MiniMax-M2.5 is cheaper at $0.30 / $1.20 per million input/output tokens, against $20 / $80 for o3-pro.
- MiniMax-M2.5 accepts more context: 205K tokens versus 200K.
- MiniMax-M2.5 has downloadable open weights; the other is API-only.
Side by side
| MiniMax-M2.5 | o3-pro | |
|---|---|---|
| Provider | MiniMax | OpenAI |
| Noometry Index | 38.3 | 42.9 |
| Released | 2026-02-12 | 2025-06-10 |
| Weights | Open | Proprietary |
| Context window | 205K | 200K |
| Max output | 131K | 100K |
| Input $ / M tokens | $0.30 | $20 |
| Output $ / M tokens | $1.20 | $80 |
| Results tracked | 33 | 12 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding o3-pro leads
MiniMax-M2.5: 48.1 (#58), o3-pro: 55.5 (#24)
| Benchmark | MiniMax-M2.5 | o3-pro |
|---|---|---|
| SWE-bench Verified (bash only) | 75.8% | — |
| Aider Polyglot | — | 84.9% |
| LMArena WebDev | 1387 | — |
| SWE-bench Multilingual | 68.3% | — |
| WeirdML | — | 58.2% |
| LMArena Coding | 1381 | — |
| ALE-Bench | 618.17 | — |
Agentic & Tool Use Not comparable
MiniMax-M2.5: 30.4 (#77), o3-pro: —
| Benchmark | MiniMax-M2.5 | o3-pro |
|---|---|---|
| Terminal-Bench | 42.7% | — |
| Vending-Bench 2 | -23.16 | — |
Reasoning o3-pro leads
MiniMax-M2.5: 17.5 (#292), o3-pro: 23.8 (#171)
| Benchmark | MiniMax-M2.5 | o3-pro |
|---|---|---|
| ARC-AGI-2 | 4.9% | 4.9% |
| Kagi LLM Benchmark | 55.2% | 72.1% |
| ARC-AGI-1 | 63.7% | 59.3% |
| Epoch Capabilities Index | 146.68 | 147.42 |
| NYT Connections (extended) | 16.8% | — |
| LMArena Hard Prompts | 1372 | — |
| DTBench | — | 86.9% |
| LMCA | — | 38.5% |
Math Not comparable
MiniMax-M2.5: 26.9 (#253), o3-pro: —
| Benchmark | MiniMax-M2.5 | o3-pro |
|---|---|---|
| ProofBench | 4% | — |
| LMArena Math | 1378 | — |
Knowledge MiniMax-M2.5 leads
MiniMax-M2.5: 39.2 (#135), o3-pro: 29.5 (#238)
| Benchmark | MiniMax-M2.5 | o3-pro |
|---|---|---|
| Vectara Hallucination Rate | 9.1% | 23.3% |
| Confabulations | — | 14.2% |
| LMArena Expert | 1379 | — |
Multilingual Not comparable
MiniMax-M2.5: 47.1 (#152), o3-pro: —
| Benchmark | MiniMax-M2.5 | o3-pro |
|---|---|---|
| LMArena Non-English | 1338 | — |
| LMArena Chinese | 1393 | — |
| LMArena French | 1362 | — |
| LMArena German | 1362 | — |
| LMArena Japanese | 1171 | — |
| LMArena Korean | 1232 | — |
| LMArena Russian | 1358 | — |
| LMArena Spanish | 1354 | — |
Instruction Following Not comparable
MiniMax-M2.5: 71.5 (#148), o3-pro: —
| Benchmark | MiniMax-M2.5 | o3-pro |
|---|---|---|
| LMArena Instruction Following | 1353 | — |
Long Context o3-pro leads
MiniMax-M2.5: 37.5 (#216), o3-pro: 72.2 (#1)
| Benchmark | MiniMax-M2.5 | o3-pro |
|---|---|---|
| Fiction.LiveBench | — | 97.2% |
| CL-bench | 11.4% | — |
| CL-bench Life | 6.3% | — |
| LMArena Longer Query | 1366 | — |
Writing & Preference o3-pro leads
MiniMax-M2.5: 53.9 (#153), o3-pro: 57.1 (#133)
| Benchmark | MiniMax-M2.5 | o3-pro |
|---|---|---|
| LMArena Text | 1359 | — |
| LMArena Creative Writing | 1331 | — |
| Short-Story Creative Writing | — | 84.4% |
| EQ-Bench Creative Writing | 1361 | — |
| LMArena Multi-Turn | 1364 | — |
Frequently asked questions
Is MiniMax-M2.5 better than o3-pro?
o3-pro is the stronger model overall, scoring 42.9 to 38.3 on the Noometry Index. MiniMax-M2.5 costs 67× less per token, which makes it the better buy when o3-pro's lead doesn't matter for your workload.
Which is cheaper, MiniMax-M2.5 or o3-pro?
MiniMax-M2.5 is cheaper. It lists at $0.30 per million input tokens and $1.20 per million output tokens; o3-pro lists at $20 and $80.
Is MiniMax-M2.5 or o3-pro better for coding?
o3-pro scores higher on coding benchmarks: 55.5 versus 48.1 in the Noometry coding category.
Which has the bigger context window?
MiniMax-M2.5 does, with 205K tokens against 200K.
How many benchmarks do MiniMax-M2.5 and o3-pro share?
5 benchmarks have published results for both models. MiniMax-M2.5 has 33 scored results on Noometry and o3-pro has 12.