# MiniMax-M2 vs Mistral Small

> MiniMax-M2 is the stronger model overall, scoring 36.9 to 33.1 on the Noometry Index. Mistral Small costs 2.0× less per token, which makes it the better buy when MiniMax-M2's lead doesn't matter for your workload.

- Canonical page: https://noometry.com/compare/minimax-m2-vs-mistral-small
- Last updated: 2026-10-10
- Shared benchmarks: 16

## Summary

- They share 16 benchmarks with published results for both. MiniMax-M2 scores higher in 6 categories and Mistral Small in 3 categories; 5 gaps are clear of the uncertainty.
- The widest gap is in math, where MiniMax-M2 leads 34.6 to 15.4.
- The biggest single-benchmark swing is Kagi LLM Benchmark: 57.8% for MiniMax-M2 and 37.8% for Mistral Small.
- Mistral Small is cheaper at $0.15 / $0.60 per million input/output tokens, against $0.30 / $1.20 for MiniMax-M2.
- Mistral Small accepts more context: 256K tokens versus 205K.

## Snapshot

| | MiniMax-M2 | Mistral Small |
|---|---|---|
| Provider | MiniMax | Mistral AI |
| Noometry Index | 36.9 | 33.1 |
| Rank | 201 | 238 |
| Context | 205K | 256K |
| Input $/M | $0.30 | $0.15 |
| Output $/M | $1.20 | $0.60 |
| Weights | Open | Open |

## Coding

- MiniMax-M2: 39.1 (#151)
- Mistral Small: 33.5 (#242)

| Benchmark | MiniMax-M2 | Mistral Small |
|---|---|---|
| LMArena Coding | 1370 | 1363 |
| SWE-bench Verified (bash only) | 61% | — |
| LMArena WebDev | 1297 | — |
| SciCode | — | 26.5% |
| BigCodeBench Instruct | — | 36.1% |
| LiveBench Coding | — | 36.2% |
| BigCodeBench Complete | — | 46.6% |
| ALE-Bench | — | 497.62 |

## Agentic & Tool Use

- MiniMax-M2: 24.9 (#102)
- Mistral Small: 27.9 (#87)

| Benchmark | MiniMax-M2 | Mistral Small |
|---|---|---|
| Terminal-Bench | 30% | — |
| Berkeley Function Calling Leaderboard | — | 37.1% |
| Vending-Bench 2 | 160.6 | — |

## Reasoning

- MiniMax-M2: 18.6 (#256)
- Mistral Small: 19.3 (#238)

| Benchmark | MiniMax-M2 | Mistral Small |
|---|---|---|
| Kagi LLM Benchmark | 57.8% | 37.8% |
| LMArena Hard Prompts | 1357 | 1336 |
| NYT Connections (extended) | 14.8% | — |
| CritPt | — | 0% |
| LiveBench Reasoning | — | 44.8% |
| DTBench | — | 70.9% |
| LiveBench Data Analysis | — | 53.7% |
| LMCA | — | 20.6% |
| LiveBench | — | 44% |

## Math

- MiniMax-M2: 34.6 (#158)
- Mistral Small: 15.4 (#287)

| Benchmark | MiniMax-M2 | Mistral Small |
|---|---|---|
| LMArena Math | 1350 | 1341 |
| OTIS Mock AIME 2024-2025 | — | 5.8% |
| LiveBench Math | — | 39.9% |
| MATH Level 5 | — | 46.8% |

## Knowledge

- MiniMax-M2: 36.3 (#160)
- Mistral Small: 30.9 (#214)

| Benchmark | MiniMax-M2 | Mistral Small |
|---|---|---|
| LMArena Expert | 1336 | 1293 |
| GPQA Diamond | — | 47.5% |
| Vectara Hallucination Rate | — | 5.1% |
| MMLU | — | 68.7% |

## Multimodal

- MiniMax-M2: —
- Mistral Small: 32.9 (#91)

| Benchmark | MiniMax-M2 | Mistral Small |
|---|---|---|
| LMArena Vision | — | 1142 |

## Multilingual

- MiniMax-M2: 45.6 (#165)
- Mistral Small: 45.7 (#164)

| Benchmark | MiniMax-M2 | Mistral Small |
|---|---|---|
| LMArena Non-English | 1314 | 1315 |
| LMArena Chinese | 1367 | 1342 |
| LMArena French | 1336 | 1337 |
| LMArena German | 1366 | 1341 |
| LMArena Russian | 1332 | 1324 |
| LMArena Spanish | 1326 | 1346 |
| LMArena Japanese | — | 1275 |
| LMArena Korean | — | 1260 |

## Instruction Following

- MiniMax-M2: 70.4 (#160)
- Mistral Small: 66.6 (#204)

| Benchmark | MiniMax-M2 | Mistral Small |
|---|---|---|
| LMArena Instruction Following | 1329 | 1310 |
| LiveBench Instruction Following | — | 63.7% |

## Long Context

- MiniMax-M2: 40.5 (#148)
- Mistral Small: 40.4 (#151)

| Benchmark | MiniMax-M2 | Mistral Small |
|---|---|---|
| LMArena Longer Query | 1332 | 1327 |

## Writing & Preference

- MiniMax-M2: 53.3 (#157)
- Mistral Small: 52.6 (#165)

| Benchmark | MiniMax-M2 | Mistral Small |
|---|---|---|
| LMArena Text | 1340 | 1338 |
| LMArena Creative Writing | 1288 | 1304 |
| LMArena Multi-Turn | 1362 | 1344 |
| LiveBench Language | — | 30.5% |

## FAQ

### Is MiniMax-M2 better than Mistral Small?

MiniMax-M2 is the stronger model overall, scoring 36.9 to 33.1 on the Noometry Index. Mistral Small costs 2.0× less per token, which makes it the better buy when MiniMax-M2's lead doesn't matter for your workload.

### Which is cheaper, MiniMax-M2 or Mistral Small?

Mistral Small is cheaper. It lists at $0.15 per million input tokens and $0.60 per million output tokens; MiniMax-M2 lists at $0.30 and $1.20.

### Is MiniMax-M2 or Mistral Small better for coding?

MiniMax-M2 scores higher on coding benchmarks: 39.1 versus 33.5 in the Noometry coding category.

### Which has the bigger context window?

Mistral Small does, with 256K tokens against 205K.

### How many benchmarks do MiniMax-M2 and Mistral Small share?

16 benchmarks have published results for both models. MiniMax-M2 has 21 scored results on Noometry and Mistral Small has 39.