# MiniMax M1 vs MiniMax-M2.7

> MiniMax M1 is the stronger model overall, scoring 39.7 to 37.4 on the Noometry Index. MiniMax-M2.7 costs 1.6× less per token, which makes it the better buy when MiniMax M1's lead doesn't matter for your workload.

- Canonical page: https://noometry.com/compare/minimax-m1-vs-minimax-m2-7
- Last updated: 2026-10-10
- Shared benchmarks: 17

## Summary

- They share 17 benchmarks with published results for both. MiniMax M1 scores higher in 2 categories and MiniMax-M2.7 in 6 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where MiniMax M1 leads 34.8 to 24.9.
- MiniMax-M2.7 is cheaper at $0.30 / $1.20 per million input/output tokens, against $0.40 / $2.20 for MiniMax M1.
- MiniMax M1 accepts more context: 1M tokens versus 205K.

## Snapshot

| | MiniMax M1 | MiniMax-M2.7 |
|---|---|---|
| Provider | MiniMax | MiniMax |
| Noometry Index | 39.7 | 37.4 |
| Rank | 146 | 188 |
| Context | 1M | 205K |
| Input $/M | $0.40 | $0.30 |
| Output $/M | $2.20 | $1.20 |
| Weights | Open | Open |

## Coding

- MiniMax M1: 39.2 (#149)
- MiniMax-M2.7: 41.4 (#115)

| Benchmark | MiniMax M1 | MiniMax-M2.7 |
|---|---|---|
| LMArena Coding | 1359 | 1454 |
| LMArena WebDev | — | 1397 |
| SciCode | — | 47% |
| WeirdML | — | 37% |
| ALE-Bench | — | 599.25 |

## Agentic & Tool Use

- MiniMax M1: —
- MiniMax-M2.7: 24.9 (#103)

| Benchmark | MiniMax M1 | MiniMax-M2.7 |
|---|---|---|
| Terminal-Bench | — | 45.1% |
| ExploitBench | — | 13.3% |
| GBAEval | — | 0% |

## Reasoning

- MiniMax M1: 25.8 (#125)
- MiniMax-M2.7: 19.0 (#248)

| Benchmark | MiniMax M1 | MiniMax-M2.7 |
|---|---|---|
| LMArena Hard Prompts | 1340 | 1422 |
| NYT Connections (extended) | — | 24.7% |
| CritPt | — | 0.6% |
| Thematic Generalization | — | 39.3% |
| Epoch Capabilities Index | — | 145.84 |

## Math

- MiniMax M1: 34.8 (#152)
- MiniMax-M2.7: 24.9 (#254)

| Benchmark | MiniMax M1 | MiniMax-M2.7 |
|---|---|---|
| LMArena Math | 1360 | 1420 |
| ProofBench | — | 3% |

## Knowledge

- MiniMax M1: 35.7 (#168)
- MiniMax-M2.7: 37.1 (#151)

| Benchmark | MiniMax M1 | MiniMax-M2.7 |
|---|---|---|
| LMArena Expert | 1316 | 1443 |
| Vectara Hallucination Rate | — | 12.9% |

## Multilingual

- MiniMax M1: 46.0 (#157)
- MiniMax-M2.7: 50.5 (#120)

| Benchmark | MiniMax M1 | MiniMax-M2.7 |
|---|---|---|
| LMArena Non-English | 1320 | 1382 |
| LMArena Chinese | 1361 | 1439 |
| LMArena French | 1369 | 1422 |
| LMArena German | 1350 | 1398 |
| LMArena Japanese | 1220 | 1261 |
| LMArena Korean | 1263 | 1313 |
| LMArena Russian | 1330 | 1383 |
| LMArena Spanish | 1354 | 1403 |

## Instruction Following

- MiniMax M1: 69.5 (#169)
- MiniMax-M2.7: 74.2 (#101)

| Benchmark | MiniMax M1 | MiniMax-M2.7 |
|---|---|---|
| LMArena Instruction Following | 1312 | 1404 |

## Long Context

- MiniMax M1: 41.4 (#136)
- MiniMax-M2.7: 43.3 (#94)

| Benchmark | MiniMax M1 | MiniMax-M2.7 |
|---|---|---|
| LMArena Longer Query | 1326 | 1419 |
| Fiction.LiveBench | 69.4% | — |

## Writing & Preference

- MiniMax M1: 53.3 (#155)
- MiniMax-M2.7: 59.1 (#107)

| Benchmark | MiniMax M1 | MiniMax-M2.7 |
|---|---|---|
| LMArena Text | 1343 | 1404 |
| LMArena Creative Writing | 1298 | 1354 |
| LMArena Multi-Turn | 1335 | 1413 |

## FAQ

### Is MiniMax M1 better than MiniMax-M2.7?

MiniMax M1 is the stronger model overall, scoring 39.7 to 37.4 on the Noometry Index. MiniMax-M2.7 costs 1.6× less per token, which makes it the better buy when MiniMax M1's lead doesn't matter for your workload.

### Which is cheaper, MiniMax M1 or MiniMax-M2.7?

MiniMax-M2.7 is cheaper. It lists at $0.30 per million input tokens and $1.20 per million output tokens; MiniMax M1 lists at $0.40 and $2.20.

### Is MiniMax M1 or MiniMax-M2.7 better for coding?

MiniMax-M2.7 scores higher on coding benchmarks: 41.4 versus 39.2 in the Noometry coding category.

### Which has the bigger context window?

MiniMax M1 does, with 1M tokens against 205K.

### How many benchmarks do MiniMax M1 and MiniMax-M2.7 share?

17 benchmarks have published results for both models. MiniMax M1 has 18 scored results on Noometry and MiniMax-M2.7 has 30.