# MiniMax-M2.5 vs Mistral Medium

> MiniMax-M2.5 is the stronger model overall, scoring 38.1 to 36.0 on the Noometry Index.

- Canonical page: https://noometry.com/compare/minimax-m2-5-vs-mistral-medium
- Last updated: 2026-10-10
- Shared benchmarks: 21

## Summary

- They share 21 benchmarks with published results for both. MiniMax-M2.5 scores higher in 3 categories and Mistral Medium in 6 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in coding, where MiniMax-M2.5 leads 47.9 to 33.8.
- The biggest single-benchmark swing is Vectara Hallucination Rate: 9.1% for MiniMax-M2.5 and 22.7% for Mistral Medium.
- MiniMax-M2.5 is cheaper at $0.30 / $1.20 per million input/output tokens, against $1.50 / $7.50 for Mistral Medium.
- Mistral Medium accepts more context: 262K tokens versus 205K.

## Snapshot

| | MiniMax-M2.5 | Mistral Medium |
|---|---|---|
| Provider | MiniMax | Mistral AI |
| Noometry Index | 38.1 | 36.0 |
| Rank | 181 | 211 |
| Context | 205K | 262K |
| Input $/M | $0.30 | $1.50 |
| Output $/M | $1.20 | $7.50 |
| Weights | Open | Open |

## Coding

- MiniMax-M2.5: 47.9 (#53)
- Mistral Medium: 33.8 (#241)

| Benchmark | MiniMax-M2.5 | Mistral Medium |
|---|---|---|
| LMArena Coding | 1380 | 1434 |
| ALE-Bench | 618.17 | 763.98 |
| FrontierCode | — | 8% |
| SWE-bench Verified (bash only) | 75.8% | — |
| LMArena WebDev | 1386 | — |
| SWE-bench Multilingual | 68.3% | — |
| SciCode | — | 40.2% |
| WeirdML | — | 43.7% |

## Agentic & Tool Use

- MiniMax-M2.5: 30.2 (#70)
- Mistral Medium: 28.1 (#84)

| Benchmark | MiniMax-M2.5 | Mistral Medium |
|---|---|---|
| Terminal-Bench | 42.7% | — |
| Berkeley Function Calling Leaderboard | — | 37.7% |
| Vending-Bench 2 | -23.16 | — |

## Reasoning

- MiniMax-M2.5: 17.6 (#281)
- Mistral Medium: 23.3 (#165)

| Benchmark | MiniMax-M2.5 | Mistral Medium |
|---|---|---|
| Kagi LLM Benchmark | 55.2% | 50% |
| LMArena Hard Prompts | 1372 | 1426 |
| ARC-AGI-2 | 4.9% | — |
| NYT Connections (extended) | 16.8% | — |
| ARC-AGI-1 | 63.7% | — |
| CritPt | — | 0% |
| DTBench | — | 75.5% |
| LMCA | — | 26.1% |
| Surface Evolver Bench | — | 26.9% |
| Epoch Capabilities Index | 146.68 | — |

## Math

- MiniMax-M2.5: 25.8 (#242)
- Mistral Medium: 27.2 (#238)

| Benchmark | MiniMax-M2.5 | Mistral Medium |
|---|---|---|
| ProofBench | 4% | 9% |
| LMArena Math | 1378 | 1409 |
| OTIS Mock AIME 2024-2025 | — | 32.2% |
| MATH Level 5 | — | 81.6% |
| FrontierMath (Feb 2025 set) | — | 0.3% |

## Knowledge

- MiniMax-M2.5: 38.6 (#132)
- Mistral Medium: 24.9 (#260)

| Benchmark | MiniMax-M2.5 | Mistral Medium |
|---|---|---|
| Vectara Hallucination Rate | 9.1% | 22.7% |
| LMArena Expert | 1378 | 1408 |
| GPQA Diamond | — | 59.5% |
| Humanity's Last Exam | — | 4.5% |

## Multimodal

- MiniMax-M2.5: —
- Mistral Medium: 34.6 (#82)

| Benchmark | MiniMax-M2.5 | Mistral Medium |
|---|---|---|
| LMArena Vision | — | 1172 |

## Multilingual

- MiniMax-M2.5: 47.3 (#147)
- Mistral Medium: 52.3 (#86)

| Benchmark | MiniMax-M2.5 | Mistral Medium |
|---|---|---|
| LMArena Non-English | 1338 | 1408 |
| LMArena Chinese | 1392 | 1448 |
| LMArena French | 1363 | 1460 |
| LMArena German | 1361 | 1433 |
| LMArena Japanese | 1169 | 1378 |
| LMArena Korean | 1232 | 1382 |
| LMArena Russian | 1359 | 1411 |
| LMArena Spanish | 1354 | 1433 |

## Instruction Following

- MiniMax-M2.5: 71.7 (#143)
- Mistral Medium: 73.9 (#111)

| Benchmark | MiniMax-M2.5 | Mistral Medium |
|---|---|---|
| LMArena Instruction Following | 1353 | 1398 |

## Long Context

- MiniMax-M2.5: 37.4 (#211)
- Mistral Medium: 42.9 (#108)

| Benchmark | MiniMax-M2.5 | Mistral Medium |
|---|---|---|
| LMArena Longer Query | 1367 | 1406 |
| CL-bench | 11.4% | — |
| CL-bench Life | 6.3% | — |

## Writing & Preference

- MiniMax-M2.5: 54.1 (#149)
- Mistral Medium: 60.2 (#97)

| Benchmark | MiniMax-M2.5 | Mistral Medium |
|---|---|---|
| LMArena Text | 1359 | 1424 |
| LMArena Creative Writing | 1331 | 1391 |
| LMArena Multi-Turn | 1364 | 1418 |
| Short-Story Creative Writing | — | 77.3% |
| EQ-Bench Creative Writing | 1361 | — |

## FAQ

### Is MiniMax-M2.5 better than Mistral Medium?

MiniMax-M2.5 is the stronger model overall, scoring 38.1 to 36.0 on the Noometry Index.

### Which is cheaper, MiniMax-M2.5 or Mistral Medium?

MiniMax-M2.5 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-M2.5 or Mistral Medium better for coding?

MiniMax-M2.5 scores higher on coding benchmarks: 47.9 versus 33.8 in the Noometry coding category.

### Which has the bigger context window?

Mistral Medium does, with 262K tokens against 205K.

### How many benchmarks do MiniMax-M2.5 and Mistral Medium share?

21 benchmarks have published results for both models. MiniMax-M2.5 has 33 scored results on Noometry and Mistral Medium has 36.