Model comparison
MiniMax-M3 vs Qwen2.5 32B Instruct
MiniMax-M3 is the stronger model overall, scoring 43.8 to 30.1 on the Noometry Index.
Last verified . 4 shared benchmarks.
MiniMax
Alibaba (Qwen)Summary
- They share 4 benchmarks with published results for both. MiniMax-M3 scores higher in 4 categories and Qwen2.5 32B Instruct in 0 categories; 4 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where MiniMax-M3 leads 58.4 to 24.9.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 71.1% for MiniMax-M3 and 7.4% for Qwen2.5 32B Instruct.
- MiniMax-M3 is cheaper at $0.30 / $1.20 per million input/output tokens, against $0.70 / $2.80 for Qwen2.5 32B Instruct.
- MiniMax-M3 accepts more context: 1M tokens versus 131K.
Side by side
| MiniMax-M3 | Qwen2.5 32B Instruct | |
|---|---|---|
| Provider | MiniMax | Alibaba (Qwen) |
| Noometry Index | 43.8 | 30.1 |
| Released | 2026-06-01 | 2024-09 |
| Weights | Open | Open |
| Context window | 1M | 131K |
| Max output | 512K | 8K |
| Input $ / M tokens | $0.30 | $0.70 |
| Output $ / M tokens | $1.20 | $2.80 |
| Results tracked | 41 | 7 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding MiniMax-M3 leads
MiniMax-M3: 41.8 (#118), Qwen2.5 32B Instruct: 38.7 (#169)
| Benchmark | MiniMax-M3 | Qwen2.5 32B Instruct |
|---|---|---|
| FrontierCode | 14.7% | — |
| LMArena WebDev | 1482 | — |
| SciCode | 47.1% | — |
| BigCodeBench Instruct | — | 45% |
| LMArena Coding | 1469 | — |
| BigCodeBench Complete | — | 52.3% |
| ALE-Bench | 640.02 | — |
Agentic & Tool Use Not comparable
MiniMax-M3: 22.6 (#130), Qwen2.5 32B Instruct: —
| Benchmark | MiniMax-M3 | Qwen2.5 32B Instruct |
|---|---|---|
| APEX-Agents | 37.7% | — |
| OSWorld 2.0 | 4.6% | — |
| GBAEval | 0.9% | — |
| Vending-Bench 2 | 2,158 | — |
Reasoning MiniMax-M3 leads
MiniMax-M3: 30.1 (#87), Qwen2.5 32B Instruct: 19.2 (#266)
| Benchmark | MiniMax-M3 | Qwen2.5 32B Instruct |
|---|---|---|
| Chess Puzzles | 14% | 0% |
| Epoch Capabilities Index | 146.95 | 128.52 |
| SimpleBench | 45.8% | — |
| NYT Connections (extended) | 65.1% | — |
| CritPt | 3.7% | — |
| LMArena Hard Prompts | 1447 | — |
| Mystery Game Puzzles | 8% | — |
| DTBench | 78.9% | — |
| LMCA | 33.7% | — |
| Surface Evolver Bench | 55% | — |
| ForecastBench | 61.4 | — |
Math MiniMax-M3 leads
MiniMax-M3: 40.0 (#95), Qwen2.5 32B Instruct: 16.2 (#296)
| Benchmark | MiniMax-M3 | Qwen2.5 32B Instruct |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 71.1% | 7.4% |
| ProofBench | 18% | — |
| LMArena Math | 1429 | — |
| MATH Level 5 | — | 56.1% |
Knowledge MiniMax-M3 leads
MiniMax-M3: 58.4 (#35), Qwen2.5 32B Instruct: 24.9 (#266)
| Benchmark | MiniMax-M3 | Qwen2.5 32B Instruct |
|---|---|---|
| GPQA Diamond | 90.9% | 46.1% |
| LMArena Expert | 1461 | — |
Multimodal Not comparable
MiniMax-M3: 40.2 (#51), Qwen2.5 32B Instruct: —
| Benchmark | MiniMax-M3 | Qwen2.5 32B Instruct |
|---|---|---|
| LMArena Vision | 1253 | — |
| LMArena Document | 1435 | — |
Multilingual Not comparable
MiniMax-M3: 53.0 (#75), Qwen2.5 32B Instruct: —
| Benchmark | MiniMax-M3 | Qwen2.5 32B Instruct |
|---|---|---|
| LMArena Non-English | 1420 | — |
| LMArena Chinese | 1463 | — |
| LMArena French | 1447 | — |
| LMArena German | 1426 | — |
| LMArena Japanese | 1381 | — |
| LMArena Korean | 1372 | — |
| LMArena Russian | 1428 | — |
| LMArena Spanish | 1432 | — |
Instruction Following Not comparable
MiniMax-M3: 75.5 (#62), Qwen2.5 32B Instruct: —
| Benchmark | MiniMax-M3 | Qwen2.5 32B Instruct |
|---|---|---|
| LMArena Instruction Following | 1433 | — |
Long Context Not comparable
MiniMax-M3: 44.2 (#72), Qwen2.5 32B Instruct: —
| Benchmark | MiniMax-M3 | Qwen2.5 32B Instruct |
|---|---|---|
| LMArena Longer Query | 1445 | — |
Writing & Preference Not comparable
MiniMax-M3: 62.1 (#83), Qwen2.5 32B Instruct: —
| Benchmark | MiniMax-M3 | Qwen2.5 32B Instruct |
|---|---|---|
| LMArena Text | 1433 | — |
| LMArena Creative Writing | 1404 | — |
| EQ-Bench 4 | 1150 | — |
| LMArena Multi-Turn | 1442 | — |
Frequently asked questions
Is MiniMax-M3 better than Qwen2.5 32B Instruct?
MiniMax-M3 is the stronger model overall, scoring 43.8 to 30.1 on the Noometry Index.
Which is cheaper, MiniMax-M3 or Qwen2.5 32B Instruct?
MiniMax-M3 is cheaper. It lists at $0.30 per million input tokens and $1.20 per million output tokens; Qwen2.5 32B Instruct lists at $0.70 and $2.80.
Is MiniMax-M3 or Qwen2.5 32B Instruct better for coding?
MiniMax-M3 scores higher on coding benchmarks: 41.8 versus 38.7 in the Noometry coding category.
Which has the bigger context window?
MiniMax-M3 does, with 1M tokens against 131K.
How many benchmarks do MiniMax-M3 and Qwen2.5 32B Instruct share?
4 benchmarks have published results for both models. MiniMax-M3 has 41 scored results on Noometry and Qwen2.5 32B Instruct has 7.