Model comparison
MiniMax-M3 vs Mistral Medium 3.5
MiniMax-M3 is the stronger model overall, scoring 43.8 to 40.2 on the Noometry Index.
Last verified . 21 shared benchmarks.
MiniMax
Mistral AISummary
- They share 21 benchmarks with published results for both. MiniMax-M3 scores higher in 9 categories and Mistral Medium 3.5 in 0 categories; 6 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where MiniMax-M3 leads 58.4 to 40.0.
- The biggest single-benchmark swing is NYT Connections (extended): 65.1% for MiniMax-M3 and 12.9% for Mistral Medium 3.5.
- MiniMax-M3 is cheaper at $0.30 / $1.20 per million input/output tokens, against $1.50 / $7.50 for Mistral Medium 3.5.
- MiniMax-M3 accepts more context: 1M tokens versus 262K.
Side by side
| MiniMax-M3 | Mistral Medium 3.5 | |
|---|---|---|
| Provider | MiniMax | Mistral AI |
| Noometry Index | 43.8 | 40.2 |
| Released | 2026-06-01 | — |
| Weights | Open | Open |
| Context window | 1M | 262K |
| Max output | 512K | 210K |
| Input $ / M tokens | $0.30 | $1.50 |
| Output $ / M tokens | $1.20 | $7.50 |
| Results tracked | 41 | 22 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding MiniMax-M3 leads
MiniMax-M3: 41.8 (#118), Mistral Medium 3.5: 36.0 (#213)
| Benchmark | MiniMax-M3 | Mistral Medium 3.5 |
|---|---|---|
| LMArena WebDev | 1482 | 1264 |
| LMArena Coding | 1469 | 1461 |
| FrontierCode | 14.7% | — |
| SciCode | 47.1% | — |
| ALE-Bench | 640.02 | — |
Agentic & Tool Use Not comparable
MiniMax-M3: 22.6 (#130), Mistral Medium 3.5: —
| Benchmark | MiniMax-M3 | Mistral Medium 3.5 |
|---|---|---|
| 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), Mistral Medium 3.5: 17.3 (#295)
| Benchmark | MiniMax-M3 | Mistral Medium 3.5 |
|---|---|---|
| NYT Connections (extended) | 65.1% | 12.9% |
| LMArena Hard Prompts | 1447 | 1436 |
| Epoch Capabilities Index | 146.95 | 141.35 |
| SimpleBench | 45.8% | — |
| Kagi LLM Benchmark | — | 41.4% |
| CritPt | 3.7% | — |
| Chess Puzzles | 14% | — |
| Mystery Game Puzzles | 8% | — |
| DTBench | 78.9% | — |
| LMCA | 33.7% | — |
| Surface Evolver Bench | 55% | — |
| ForecastBench | 61.4 | — |
Math Too close to call
MiniMax-M3: 40.0 (#95), Mistral Medium 3.5: 39.1 (#113)
| Benchmark | MiniMax-M3 | Mistral Medium 3.5 |
|---|---|---|
| LMArena Math | 1429 | 1431 |
| OTIS Mock AIME 2024-2025 | 71.1% | — |
| ProofBench | 18% | — |
Knowledge MiniMax-M3 leads
MiniMax-M3: 58.4 (#35), Mistral Medium 3.5: 40.0 (#126)
| Benchmark | MiniMax-M3 | Mistral Medium 3.5 |
|---|---|---|
| LMArena Expert | 1461 | 1432 |
| GPQA Diamond | 90.9% | — |
Multimodal MiniMax-M3 leads
MiniMax-M3: 40.2 (#51), Mistral Medium 3.5: 38.3 (#65)
| Benchmark | MiniMax-M3 | Mistral Medium 3.5 |
|---|---|---|
| LMArena Vision | 1253 | 1223 |
| LMArena Document | 1435 | — |
Multilingual MiniMax-M3 leads
MiniMax-M3: 53.0 (#75), Mistral Medium 3.5: 51.9 (#100)
| Benchmark | MiniMax-M3 | Mistral Medium 3.5 |
|---|---|---|
| LMArena Non-English | 1420 | 1404 |
| LMArena Chinese | 1463 | 1442 |
| LMArena French | 1447 | 1448 |
| LMArena German | 1426 | 1451 |
| LMArena Korean | 1372 | 1385 |
| LMArena Russian | 1428 | 1395 |
| LMArena Spanish | 1432 | 1409 |
| LMArena Japanese | 1381 | — |
Instruction Following Too close to call
MiniMax-M3: 75.5 (#62), Mistral Medium 3.5: 74.6 (#90)
| Benchmark | MiniMax-M3 | Mistral Medium 3.5 |
|---|---|---|
| LMArena Instruction Following | 1433 | 1415 |
Long Context Too close to call
MiniMax-M3: 44.2 (#72), Mistral Medium 3.5: 43.2 (#103)
| Benchmark | MiniMax-M3 | Mistral Medium 3.5 |
|---|---|---|
| LMArena Longer Query | 1445 | 1415 |
Writing & Preference MiniMax-M3 leads
MiniMax-M3: 62.1 (#83), Mistral Medium 3.5: 58.5 (#117)
| Benchmark | MiniMax-M3 | Mistral Medium 3.5 |
|---|---|---|
| LMArena Text | 1433 | 1421 |
| LMArena Creative Writing | 1404 | 1374 |
| EQ-Bench 4 | 1150 | 993 |
| LMArena Multi-Turn | 1442 | 1423 |
Frequently asked questions
Is MiniMax-M3 better than Mistral Medium 3.5?
MiniMax-M3 is the stronger model overall, scoring 43.8 to 40.2 on the Noometry Index.
Which is cheaper, MiniMax-M3 or Mistral Medium 3.5?
MiniMax-M3 is cheaper. It lists at $0.30 per million input tokens and $1.20 per million output tokens; Mistral Medium 3.5 lists at $1.50 and $7.50.
Is MiniMax-M3 or Mistral Medium 3.5 better for coding?
MiniMax-M3 scores higher on coding benchmarks: 41.8 versus 36.0 in the Noometry coding category.
Which has the bigger context window?
MiniMax-M3 does, with 1M tokens against 262K.
How many benchmarks do MiniMax-M3 and Mistral Medium 3.5 share?
21 benchmarks have published results for both models. MiniMax-M3 has 41 scored results on Noometry and Mistral Medium 3.5 has 22.