# Magistral Small vs MiniMax-M2.5

> MiniMax-M2.5 is the stronger model overall, scoring 38.1 to 29.9 on the Noometry Index.

- Canonical page: https://noometry.com/compare/magistral-small-vs-minimax-m2-5
- Last updated: 2026-10-10
- Shared benchmarks: 4

## Summary

- They share 4 benchmarks with published results for both. Magistral Small scores higher in 0 categories and MiniMax-M2.5 in 4 categories; 3 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where MiniMax-M2.5 leads 17.6 to 6.9.
- The biggest single-benchmark swing is ARC-AGI-1: 5% for Magistral Small and 63.7% for MiniMax-M2.5.
- MiniMax-M2.5 is cheaper at $0.30 / $1.20 per million input/output tokens, against $0.50 / $1.50 for Magistral Small.
- MiniMax-M2.5 accepts more context: 205K tokens versus 128K.

## Snapshot

| | Magistral Small | MiniMax-M2.5 |
|---|---|---|
| Provider | Mistral AI | MiniMax |
| Noometry Index | 29.9 | 38.1 |
| Rank | 289 | 181 |
| Context | 128K | 205K |
| Input $/M | $0.50 | $0.30 |
| Output $/M | $1.50 | $1.20 |
| Weights | Open | Open |

## Coding

- Magistral Small: 37.8 (#172)
- MiniMax-M2.5: 47.9 (#53)

| Benchmark | Magistral Small | MiniMax-M2.5 |
|---|---|---|
| SWE-bench Verified (bash only) | — | 75.8% |
| LMArena WebDev | — | 1386 |
| SWE-bench Multilingual | — | 68.3% |
| SciCode | 35.2% | — |
| LMArena Coding | — | 1380 |
| ALE-Bench | — | 618.17 |

## Agentic & Tool Use

- Magistral Small: —
- MiniMax-M2.5: 30.2 (#70)

| Benchmark | Magistral Small | MiniMax-M2.5 |
|---|---|---|
| Terminal-Bench | — | 42.7% |
| Vending-Bench 2 | — | -23.16 |

## Reasoning

- Magistral Small: 6.9 (#345)
- MiniMax-M2.5: 17.6 (#281)

| Benchmark | Magistral Small | MiniMax-M2.5 |
|---|---|---|
| ARC-AGI-2 | 0% | 4.9% |
| Kagi LLM Benchmark | 6.3% | 55.2% |
| ARC-AGI-1 | 5% | 63.7% |
| Epoch Capabilities Index | 133.19 | 146.68 |
| NYT Connections (extended) | — | 16.8% |
| CritPt | 0.3% | — |
| Chess Puzzles | 3% | — |
| LMArena Hard Prompts | — | 1372 |
| DTBench | 61.3% | — |

## Math

- Magistral Small: 24.9 (#253)
- MiniMax-M2.5: 25.8 (#242)

| Benchmark | Magistral Small | MiniMax-M2.5 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 30% | — |
| ProofBench | — | 4% |
| LMArena Math | — | 1378 |

## Knowledge

- Magistral Small: 30.9 (#215)
- MiniMax-M2.5: 38.6 (#132)

| Benchmark | Magistral Small | MiniMax-M2.5 |
|---|---|---|
| GPQA Diamond | 56.1% | — |
| Vectara Hallucination Rate | — | 9.1% |
| LMArena Expert | — | 1378 |

## Multilingual

- Magistral Small: —
- MiniMax-M2.5: 47.3 (#147)

| Benchmark | Magistral Small | MiniMax-M2.5 |
|---|---|---|
| LMArena Non-English | — | 1338 |
| LMArena Chinese | — | 1392 |
| LMArena French | — | 1363 |
| LMArena German | — | 1361 |
| LMArena Japanese | — | 1169 |
| LMArena Korean | — | 1232 |
| LMArena Russian | — | 1359 |
| LMArena Spanish | — | 1354 |

## Instruction Following

- Magistral Small: —
- MiniMax-M2.5: 71.7 (#143)

| Benchmark | Magistral Small | MiniMax-M2.5 |
|---|---|---|
| LMArena Instruction Following | — | 1353 |

## Long Context

- Magistral Small: —
- MiniMax-M2.5: 37.4 (#211)

| Benchmark | Magistral Small | MiniMax-M2.5 |
|---|---|---|
| CL-bench | — | 11.4% |
| CL-bench Life | — | 6.3% |
| LMArena Longer Query | — | 1367 |

## Writing & Preference

- Magistral Small: —
- MiniMax-M2.5: 54.1 (#149)

| Benchmark | Magistral Small | MiniMax-M2.5 |
|---|---|---|
| LMArena Text | — | 1359 |
| LMArena Creative Writing | — | 1331 |
| EQ-Bench Creative Writing | — | 1361 |
| LMArena Multi-Turn | — | 1364 |

## FAQ

### Is Magistral Small better than MiniMax-M2.5?

MiniMax-M2.5 is the stronger model overall, scoring 38.1 to 29.9 on the Noometry Index.

### Which is cheaper, Magistral Small or MiniMax-M2.5?

MiniMax-M2.5 is cheaper. It lists at $0.30 per million input tokens and $1.20 per million output tokens; Magistral Small lists at $0.50 and $1.50.

### Is Magistral Small or MiniMax-M2.5 better for coding?

MiniMax-M2.5 scores higher on coding benchmarks: 47.9 versus 37.8 in the Noometry coding category.

### Which has the bigger context window?

MiniMax-M2.5 does, with 205K tokens against 128K.

### How many benchmarks do Magistral Small and MiniMax-M2.5 share?

4 benchmarks have published results for both models. Magistral Small has 10 scored results on Noometry and MiniMax-M2.5 has 33.