# MiniMax-M3 vs Mistral Medium

> MiniMax-M3 is the stronger model overall, scoring 43.8 to 36.3 on the Noometry Index.

- Canonical page: https://noometry.com/compare/minimax-m3-vs-mistral-medium
- Last updated: 2026-10-11
- Shared benchmarks: 28

## Summary

- They share 28 benchmarks with published results for both. MiniMax-M3 scores higher in 9 categories and Mistral Medium in 1 category; 9 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where MiniMax-M3 leads 58.4 to 25.0.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 71.1% for MiniMax-M3 and 32.2% for Mistral Medium.
- MiniMax-M3 is cheaper at $0.30 / $1.20 per million input/output tokens, against $1.50 / $7.50 for Mistral Medium.
- MiniMax-M3 accepts more context: 1M tokens versus 262K.

## Snapshot

| | MiniMax-M3 | Mistral Medium |
|---|---|---|
| Provider | MiniMax | Mistral AI |
| Noometry Index | 43.8 | 36.3 |
| Rank | 85 | 218 |
| Context | 1M | 262K |
| Input $/M | $0.30 | $1.50 |
| Output $/M | $1.20 | $7.50 |
| Weights | Open | Open |

## Coding

- MiniMax-M3: 41.8 (#118)
- Mistral Medium: 34.2 (#243)

| Benchmark | MiniMax-M3 | Mistral Medium |
|---|---|---|
| FrontierCode | 14.7% | 8% |
| SciCode | 47.1% | 40.2% |
| LMArena Coding | 1469 | 1434 |
| ALE-Bench | 640.02 | 763.98 |
| LMArena WebDev | 1482 | — |
| WeirdML | — | 43.7% |

## Agentic & Tool Use

- MiniMax-M3: 22.6 (#130)
- Mistral Medium: 28.3 (#90)

| Benchmark | MiniMax-M3 | Mistral Medium |
|---|---|---|
| APEX-Agents | 37.7% | — |
| Berkeley Function Calling Leaderboard | — | 37.7% |
| OSWorld 2.0 | 4.6% | — |
| GBAEval | 0.9% | — |
| Vending-Bench 2 | 2,158 | — |

## Reasoning

- MiniMax-M3: 30.1 (#87)
- Mistral Medium: 24.0 (#167)

| Benchmark | MiniMax-M3 | Mistral Medium |
|---|---|---|
| CritPt | 3.7% | 0% |
| LMArena Hard Prompts | 1447 | 1426 |
| DTBench | 78.9% | 75.5% |
| LMCA | 33.7% | 26.1% |
| Surface Evolver Bench | 55% | 26.9% |
| SimpleBench | 45.8% | — |
| Kagi LLM Benchmark | — | 50% |
| NYT Connections (extended) | 65.1% | — |
| Chess Puzzles | 14% | — |
| Mystery Game Puzzles | 8% | — |
| Epoch Capabilities Index | 146.95 | — |
| ForecastBench | 61.4 | — |

## Math

- MiniMax-M3: 40.0 (#95)
- Mistral Medium: 28.1 (#245)

| Benchmark | MiniMax-M3 | Mistral Medium |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 71.1% | 32.2% |
| ProofBench | 18% | 9% |
| LMArena Math | 1429 | 1408 |
| MATH Level 5 | — | 81.6% |
| FrontierMath (Feb 2025 set) | — | 0.3% |

## Knowledge

- MiniMax-M3: 58.4 (#35)
- Mistral Medium: 25.0 (#265)

| Benchmark | MiniMax-M3 | Mistral Medium |
|---|---|---|
| GPQA Diamond | 90.9% | 59.5% |
| LMArena Expert | 1461 | 1408 |
| Humanity's Last Exam | — | 4.5% |
| Vectara Hallucination Rate | — | 22.7% |

## Multimodal

- MiniMax-M3: 40.2 (#51)
- Mistral Medium: 35.3 (#88)

| Benchmark | MiniMax-M3 | Mistral Medium |
|---|---|---|
| LMArena Vision | 1253 | 1172 |
| LMArena Document | 1435 | — |

## Multilingual

- MiniMax-M3: 53.0 (#75)
- Mistral Medium: 52.1 (#91)

| Benchmark | MiniMax-M3 | Mistral Medium |
|---|---|---|
| LMArena Non-English | 1420 | 1408 |
| LMArena Chinese | 1463 | 1447 |
| LMArena French | 1447 | 1459 |
| LMArena German | 1426 | 1432 |
| LMArena Japanese | 1381 | 1378 |
| LMArena Korean | 1372 | 1380 |
| LMArena Russian | 1428 | 1411 |
| LMArena Spanish | 1432 | 1433 |

## Instruction Following

- MiniMax-M3: 75.5 (#62)
- Mistral Medium: 73.7 (#116)

| Benchmark | MiniMax-M3 | Mistral Medium |
|---|---|---|
| LMArena Instruction Following | 1433 | 1398 |

## Long Context

- MiniMax-M3: 44.2 (#72)
- Mistral Medium: 42.9 (#114)

| Benchmark | MiniMax-M3 | Mistral Medium |
|---|---|---|
| LMArena Longer Query | 1445 | 1406 |

## Writing & Preference

- MiniMax-M3: 62.1 (#83)
- Mistral Medium: 60.0 (#103)

| Benchmark | MiniMax-M3 | Mistral Medium |
|---|---|---|
| LMArena Text | 1433 | 1424 |
| LMArena Creative Writing | 1404 | 1391 |
| LMArena Multi-Turn | 1442 | 1418 |
| Short-Story Creative Writing | — | 77.3% |
| EQ-Bench 4 | 1150 | — |

## FAQ

### Is MiniMax-M3 better than Mistral Medium?

MiniMax-M3 is the stronger model overall, scoring 43.8 to 36.3 on the Noometry Index.

### Which is cheaper, MiniMax-M3 or Mistral Medium?

MiniMax-M3 is cheaper. It lists at $0.30 per million input tokens and $1.20 per million output tokens; Mistral Medium lists at $1.50 and $7.50.

### Is MiniMax-M3 or Mistral Medium better for coding?

MiniMax-M3 scores higher on coding benchmarks: 41.8 versus 34.2 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 share?

28 benchmarks have published results for both models. MiniMax-M3 has 41 scored results on Noometry and Mistral Medium has 36.