# Mistral Medium vs Qwen3.5 397B-A17B

> Qwen3.5 397B-A17B is the stronger model overall, scoring 46.0 to 36.3 on the Noometry Index.

- Canonical page: https://noometry.com/compare/mistral-medium-vs-qwen3-5-397b-a17b
- Last updated: 2026-10-12
- Shared benchmarks: 23

## Summary

- They share 23 benchmarks with published results for both. Mistral Medium scores higher in 0 categories and Qwen3.5 397B-A17B in 10 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Qwen3.5 397B-A17B leads 53.3 to 25.0.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 32.2% for Mistral Medium and 88.9% for Qwen3.5 397B-A17B.
- Qwen3.5 397B-A17B is cheaper at $0.60 / $3.60 per million input/output tokens, against $1.50 / $7.50 for Mistral Medium.

## Snapshot

| | Mistral Medium | Qwen3.5 397B-A17B |
|---|---|---|
| Provider | Mistral AI | Alibaba (Qwen) |
| Noometry Index | 36.3 | 46.0 |
| Rank | 218 | 67 |
| Context | 262K | 262K |
| Input $/M | $1.50 | $0.60 |
| Output $/M | $7.50 | $3.60 |
| Weights | Open | Open |

## Coding

- Mistral Medium: 34.2 (#243)
- Qwen3.5 397B-A17B: 42.0 (#114)

| Benchmark | Mistral Medium | Qwen3.5 397B-A17B |
|---|---|---|
| LMArena Coding | 1434 | 1465 |
| FrontierCode | 8% | — |
| LMArena WebDev | — | 1400 |
| SciCode | 40.2% | — |
| WeirdML | 43.7% | — |
| ALE-Bench | 763.98 | — |

## Agentic & Tool Use

- Mistral Medium: 28.3 (#90)
- Qwen3.5 397B-A17B: 33.3 (#53)

| Benchmark | Mistral Medium | Qwen3.5 397B-A17B |
|---|---|---|
| APEX-Agents | — | 24.9% |
| Berkeley Function Calling Leaderboard | 37.7% | — |
| τ²-bench Airline | — | 81.5% |
| τ²-bench Banking | — | 9.8% |
| τ²-bench Retail | — | 84.4% |
| τ²-bench Telecom | — | 97.8% |

## Reasoning

- Mistral Medium: 24.0 (#167)
- Qwen3.5 397B-A17B: 34.5 (#70)

| Benchmark | Mistral Medium | Qwen3.5 397B-A17B |
|---|---|---|
| Kagi LLM Benchmark | 50% | 73.7% |
| LMArena Hard Prompts | 1426 | 1448 |
| DTBench | 75.5% | 87.5% |
| LMCA | 26.1% | 37.9% |
| NYT Connections (extended) | — | 58.9% |
| CritPt | 0% | — |
| Chess Puzzles | — | 13% |
| Thematic Generalization | — | 65.1% |
| Mystery Game Puzzles | — | 18% |
| Surface Evolver Bench | 26.9% | — |
| Epoch Capabilities Index | — | 146.65 |

## Math

- Mistral Medium: 28.1 (#245)
- Qwen3.5 397B-A17B: 46.1 (#73)

| Benchmark | Mistral Medium | Qwen3.5 397B-A17B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 32.2% | 88.9% |
| LMArena Math | 1408 | 1454 |
| FrontierMath (Tiers 1-3) | — | 31.2% |
| ProofBench | 9% | — |
| MATH Level 5 | 81.6% | — |
| FrontierMath (Feb 2025 set) | 0.3% | — |

## Knowledge

- Mistral Medium: 25.0 (#265)
- Qwen3.5 397B-A17B: 53.3 (#58)

| Benchmark | Mistral Medium | Qwen3.5 397B-A17B |
|---|---|---|
| GPQA Diamond | 59.5% | 86.4% |
| LMArena Expert | 1408 | 1462 |
| Humanity's Last Exam | 4.5% | — |
| Vectara Hallucination Rate | 22.7% | — |

## Multimodal

- Mistral Medium: 35.3 (#88)
- Qwen3.5 397B-A17B: 40.7 (#44)

| Benchmark | Mistral Medium | Qwen3.5 397B-A17B |
|---|---|---|
| LMArena Vision | 1172 | 1263 |

## Multilingual

- Mistral Medium: 52.1 (#91)
- Qwen3.5 397B-A17B: 53.7 (#59)

| Benchmark | Mistral Medium | Qwen3.5 397B-A17B |
|---|---|---|
| LMArena Non-English | 1408 | 1430 |
| LMArena Chinese | 1447 | 1500 |
| LMArena French | 1459 | 1461 |
| LMArena German | 1432 | 1447 |
| LMArena Japanese | 1378 | 1426 |
| LMArena Korean | 1380 | 1384 |
| LMArena Russian | 1411 | 1429 |
| LMArena Spanish | 1433 | 1441 |

## Instruction Following

- Mistral Medium: 73.7 (#116)
- Qwen3.5 397B-A17B: 75.0 (#77)

| Benchmark | Mistral Medium | Qwen3.5 397B-A17B |
|---|---|---|
| LMArena Instruction Following | 1398 | 1424 |

## Long Context

- Mistral Medium: 42.9 (#114)
- Qwen3.5 397B-A17B: 44.1 (#74)

| Benchmark | Mistral Medium | Qwen3.5 397B-A17B |
|---|---|---|
| LMArena Longer Query | 1406 | 1442 |

## Writing & Preference

- Mistral Medium: 60.0 (#103)
- Qwen3.5 397B-A17B: 62.3 (#79)

| Benchmark | Mistral Medium | Qwen3.5 397B-A17B |
|---|---|---|
| LMArena Text | 1424 | 1438 |
| LMArena Creative Writing | 1391 | 1401 |
| LMArena Multi-Turn | 1418 | 1446 |
| Short-Story Creative Writing | 77.3% | — |
| EQ-Bench Creative Writing | — | 1478 |

## FAQ

### Is Mistral Medium better than Qwen3.5 397B-A17B?

Qwen3.5 397B-A17B is the stronger model overall, scoring 46.0 to 36.3 on the Noometry Index.

### Which is cheaper, Mistral Medium or Qwen3.5 397B-A17B?

Qwen3.5 397B-A17B is cheaper. It lists at $0.60 per million input tokens and $3.60 per million output tokens; Mistral Medium lists at $1.50 and $7.50.

### Is Mistral Medium or Qwen3.5 397B-A17B better for coding?

Qwen3.5 397B-A17B scores higher on coding benchmarks: 42.0 versus 34.2 in the Noometry coding category.

### Which has the bigger context window?

Both accept 262K tokens.

### How many benchmarks do Mistral Medium and Qwen3.5 397B-A17B share?

23 benchmarks have published results for both models. Mistral Medium has 36 scored results on Noometry and Qwen3.5 397B-A17B has 36.