# MiniMax-M2.5 vs Qwen3.6 Plus

> Qwen3.6 Plus is the stronger model overall, scoring 47.3 to 38.1 on the Noometry Index. MiniMax-M2.5 costs 2.1× less per token, which makes it the better buy when Qwen3.6 Plus's lead doesn't matter for your workload.

- Canonical page: https://noometry.com/compare/minimax-m2-5-vs-qwen3-6-plus
- Last updated: 2026-10-10
- Shared benchmarks: 23

## Summary

- They share 23 benchmarks with published results for both. MiniMax-M2.5 scores higher in 1 category and Qwen3.6 Plus in 7 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Qwen3.6 Plus leads 50.6 to 25.8.
- The biggest single-benchmark swing is NYT Connections (extended): 16.8% for MiniMax-M2.5 and 60.3% for Qwen3.6 Plus.
- MiniMax-M2.5 is cheaper at $0.30 / $1.20 per million input/output tokens, against $0.50 / $3 for Qwen3.6 Plus.
- Qwen3.6 Plus accepts more context: 1M tokens versus 205K.
- MiniMax-M2.5 has downloadable open weights; the other is API-only.

## Snapshot

| | MiniMax-M2.5 | Qwen3.6 Plus |
|---|---|---|
| Provider | MiniMax | Alibaba (Qwen) |
| Noometry Index | 38.1 | 47.3 |
| Rank | 181 | 54 |
| Context | 205K | 1M |
| Input $/M | $0.30 | $0.50 |
| Output $/M | $1.20 | $3 |
| Weights | Open | Proprietary |

## Coding

- MiniMax-M2.5: 47.9 (#53)
- Qwen3.6 Plus: 40.6 (#125)

| Benchmark | MiniMax-M2.5 | Qwen3.6 Plus |
|---|---|---|
| LMArena WebDev | 1386 | 1461 |
| LMArena Coding | 1380 | 1467 |
| ALE-Bench | 618.17 | 670.15 |
| SWE-bench Verified | — | 57.9% |
| SWE-bench Verified (bash only) | 75.8% | — |
| SWE-bench Multilingual | 68.3% | — |
| SciCode | — | 40.7% |

## Agentic & Tool Use

- MiniMax-M2.5: 30.2 (#70)
- Qwen3.6 Plus: —

| Benchmark | MiniMax-M2.5 | Qwen3.6 Plus |
|---|---|---|
| Vending-Bench 2 | -23.16 | 5,115 |
| Terminal-Bench | 42.7% | — |

## Reasoning

- MiniMax-M2.5: 17.6 (#281)
- Qwen3.6 Plus: 28.5 (#89)

| Benchmark | MiniMax-M2.5 | Qwen3.6 Plus |
|---|---|---|
| NYT Connections (extended) | 16.8% | 60.3% |
| LMArena Hard Prompts | 1372 | 1449 |
| Epoch Capabilities Index | 146.68 | 147.67 |
| ARC-AGI-2 | 4.9% | — |
| Kagi LLM Benchmark | 55.2% | — |
| ARC-AGI-1 | 63.7% | — |
| CritPt | — | 2.9% |
| Chess Puzzles | — | 17% |
| Thematic Generalization | — | 59.5% |
| Mystery Game Puzzles | — | 12% |
| DTBench | — | 81.9% |
| LMCA | — | 33.1% |

## Math

- MiniMax-M2.5: 25.8 (#242)
- Qwen3.6 Plus: 50.6 (#44)

| Benchmark | MiniMax-M2.5 | Qwen3.6 Plus |
|---|---|---|
| LMArena Math | 1378 | 1450 |
| FrontierMath (Tiers 1-3) | — | 38.2% |
| OTIS Mock AIME 2024-2025 | — | 93.3% |
| ProofBench | 4% | — |
| FrontierMath (Feb 2025 set) | — | 26.2% |
| FrontierMath Tier 4 (v1) | — | 8.3% |

## Knowledge

- MiniMax-M2.5: 38.6 (#132)
- Qwen3.6 Plus: 56.1 (#40)

| Benchmark | MiniMax-M2.5 | Qwen3.6 Plus |
|---|---|---|
| LMArena Expert | 1378 | 1456 |
| GPQA Diamond | — | 88.4% |
| SimpleQA Verified | — | 44.1% |
| Vectara Hallucination Rate | 9.1% | — |

## Multilingual

- MiniMax-M2.5: 47.3 (#147)
- Qwen3.6 Plus: 53.5 (#66)

| Benchmark | MiniMax-M2.5 | Qwen3.6 Plus |
|---|---|---|
| LMArena Non-English | 1338 | 1424 |
| LMArena Chinese | 1392 | 1478 |
| LMArena French | 1363 | 1454 |
| LMArena German | 1361 | 1451 |
| LMArena Japanese | 1169 | 1389 |
| LMArena Korean | 1232 | 1381 |
| LMArena Russian | 1359 | 1434 |
| LMArena Spanish | 1354 | 1432 |

## Instruction Following

- MiniMax-M2.5: 71.7 (#143)
- Qwen3.6 Plus: 75.2 (#70)

| Benchmark | MiniMax-M2.5 | Qwen3.6 Plus |
|---|---|---|
| LMArena Instruction Following | 1353 | 1425 |

## Long Context

- MiniMax-M2.5: 37.4 (#211)
- Qwen3.6 Plus: 45.1 (#48)

| Benchmark | MiniMax-M2.5 | Qwen3.6 Plus |
|---|---|---|
| CL-bench | 11.4% | 20.3% |
| LMArena Longer Query | 1367 | 1440 |
| CL-bench Life | 6.3% | — |

## Writing & Preference

- MiniMax-M2.5: 54.1 (#149)
- Qwen3.6 Plus: 62.4 (#78)

| Benchmark | MiniMax-M2.5 | Qwen3.6 Plus |
|---|---|---|
| LMArena Text | 1359 | 1437 |
| LMArena Creative Writing | 1331 | 1405 |
| LMArena Multi-Turn | 1364 | 1438 |
| EQ-Bench Creative Writing | 1361 | — |

## FAQ

### Is MiniMax-M2.5 better than Qwen3.6 Plus?

Qwen3.6 Plus is the stronger model overall, scoring 47.3 to 38.1 on the Noometry Index. MiniMax-M2.5 costs 2.1× less per token, which makes it the better buy when Qwen3.6 Plus's lead doesn't matter for your workload.

### Which is cheaper, MiniMax-M2.5 or Qwen3.6 Plus?

MiniMax-M2.5 is cheaper. It lists at $0.30 per million input tokens and $1.20 per million output tokens; Qwen3.6 Plus lists at $0.50 and $3.

### Is MiniMax-M2.5 or Qwen3.6 Plus better for coding?

MiniMax-M2.5 scores higher on coding benchmarks: 47.9 versus 40.6 in the Noometry coding category.

### Which has the bigger context window?

Qwen3.6 Plus does, with 1M tokens against 205K.

### How many benchmarks do MiniMax-M2.5 and Qwen3.6 Plus share?

23 benchmarks have published results for both models. MiniMax-M2.5 has 33 scored results on Noometry and Qwen3.6 Plus has 37.