# MiniMax-M3 vs Qwen3.5-Flash

> MiniMax-M3 is the stronger model overall, scoring 43.5 to 42.3 on the Noometry Index. Qwen3.5-Flash costs 3.0× less per token, which makes it the better buy when MiniMax-M3's lead doesn't matter for your workload.

- Canonical page: https://noometry.com/compare/minimax-m3-vs-qwen3-5-flash
- Last updated: 2026-10-10
- Shared benchmarks: 27

## Summary

- They share 27 benchmarks with published results for both. MiniMax-M3 scores higher in 7 categories and Qwen3.5-Flash in 1 category; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where MiniMax-M3 leads 58.4 to 43.2.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 71.1% for MiniMax-M3 and 84.4% for Qwen3.5-Flash.
- Qwen3.5-Flash is cheaper at $0.10 / $0.40 per million input/output tokens, against $0.30 / $1.20 for MiniMax-M3.
- MiniMax-M3 has downloadable open weights; the other is API-only.

## Snapshot

| | MiniMax-M3 | Qwen3.5-Flash |
|---|---|---|
| Provider | MiniMax | Alibaba (Qwen) |
| Noometry Index | 43.5 | 42.3 |
| Rank | 81 | 105 |
| Context | 1M | 1M |
| Input $/M | $0.30 | $0.10 |
| Output $/M | $1.20 | $0.40 |
| Weights | Open | Proprietary |

## Coding

- MiniMax-M3: 41.5 (#111)
- Qwen3.5-Flash: 33.9 (#234)

| Benchmark | MiniMax-M3 | Qwen3.5-Flash |
|---|---|---|
| LMArena WebDev | 1482 | 1243 |
| LMArena Coding | 1470 | 1412 |
| ALE-Bench | 640.02 | 221.8 |
| FrontierCode | 14.7% | — |
| SciCode | 47.1% | — |

## Agentic & Tool Use

- MiniMax-M3: 22.5 (#121)
- Qwen3.5-Flash: —

| Benchmark | MiniMax-M3 | Qwen3.5-Flash |
|---|---|---|
| Vending-Bench 2 | 2,158 | 462.69 |
| APEX-Agents | 37.7% | — |
| OSWorld 2.0 | 4.6% | — |
| GBAEval | 0.9% | — |

## Reasoning

- MiniMax-M3: 29.3 (#83)
- Qwen3.5-Flash: 32.9 (#63)

| Benchmark | MiniMax-M3 | Qwen3.5-Flash |
|---|---|---|
| Chess Puzzles | 14% | 21% |
| LMArena Hard Prompts | 1447 | 1404 |
| Mystery Game Puzzles | 8% | 20% |
| DTBench | 78.9% | 82.9% |
| LMCA | 33.7% | 29.1% |
| Epoch Capabilities Index | 147 | 143.98 |
| SimpleBench | 45.8% | — |
| NYT Connections (extended) | 65.1% | — |
| CritPt | 3.7% | — |
| Surface Evolver Bench | 55% | — |
| ForecastBench | 61.4 | — |

## Math

- MiniMax-M3: 39.1 (#83)
- Qwen3.5-Flash: 36.3 (#117)

| Benchmark | MiniMax-M3 | Qwen3.5-Flash |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 71.1% | 84.4% |
| LMArena Math | 1432 | 1407 |
| FrontierMath (Tiers 1-3) | — | 18.2% |
| ProofBench | 18% | — |
| FrontierMath (Feb 2025 set) | — | 6.2% |
| FrontierMath Tier 4 (v1) | — | 0% |

## Knowledge

- MiniMax-M3: 58.4 (#32)
- Qwen3.5-Flash: 43.2 (#84)

| Benchmark | MiniMax-M3 | Qwen3.5-Flash |
|---|---|---|
| GPQA Diamond | 90.9% | 82.3% |
| LMArena Expert | 1462 | 1408 |
| SimpleQA Verified | — | 20.3% |
| Vectara Hallucination Rate | — | 10.5% |

## Multimodal

- MiniMax-M3: 39.5 (#45)
- Qwen3.5-Flash: —

| Benchmark | MiniMax-M3 | Qwen3.5-Flash |
|---|---|---|
| LMArena Vision | 1254 | — |
| LMArena Document | 1435 | — |

## Multilingual

- MiniMax-M3: 53.2 (#70)
- Qwen3.5-Flash: 50.7 (#116)

| Benchmark | MiniMax-M3 | Qwen3.5-Flash |
|---|---|---|
| LMArena Non-English | 1420 | 1385 |
| LMArena Chinese | 1465 | 1446 |
| LMArena French | 1445 | 1412 |
| LMArena German | 1424 | 1389 |
| LMArena Japanese | 1381 | 1369 |
| LMArena Korean | 1372 | 1344 |
| LMArena Russian | 1430 | 1379 |
| LMArena Spanish | 1431 | 1399 |

## Instruction Following

- MiniMax-M3: 75.7 (#56)
- Qwen3.5-Flash: 72.8 (#133)

| Benchmark | MiniMax-M3 | Qwen3.5-Flash |
|---|---|---|
| LMArena Instruction Following | 1433 | 1374 |

## Long Context

- MiniMax-M3: 44.2 (#68)
- Qwen3.5-Flash: 42.4 (#119)

| Benchmark | MiniMax-M3 | Qwen3.5-Flash |
|---|---|---|
| LMArena Longer Query | 1445 | 1392 |

## Writing & Preference

- MiniMax-M3: 62.3 (#79)
- Qwen3.5-Flash: 58.2 (#117)

| Benchmark | MiniMax-M3 | Qwen3.5-Flash |
|---|---|---|
| LMArena Text | 1433 | 1397 |
| LMArena Creative Writing | 1405 | 1343 |
| LMArena Multi-Turn | 1441 | 1393 |
| EQ-Bench 4 | 1150 | — |

## FAQ

### Is MiniMax-M3 better than Qwen3.5-Flash?

MiniMax-M3 is the stronger model overall, scoring 43.5 to 42.3 on the Noometry Index. Qwen3.5-Flash costs 3.0× less per token, which makes it the better buy when MiniMax-M3's lead doesn't matter for your workload.

### Which is cheaper, MiniMax-M3 or Qwen3.5-Flash?

Qwen3.5-Flash is cheaper. It lists at $0.10 per million input tokens and $0.40 per million output tokens; MiniMax-M3 lists at $0.30 and $1.20.

### Is MiniMax-M3 or Qwen3.5-Flash better for coding?

MiniMax-M3 scores higher on coding benchmarks: 41.5 versus 33.9 in the Noometry coding category.

### Which has the bigger context window?

Both accept 1M tokens.

### How many benchmarks do MiniMax-M3 and Qwen3.5-Flash share?

27 benchmarks have published results for both models. MiniMax-M3 has 41 scored results on Noometry and Qwen3.5-Flash has 32.