# Mistral Large 3 vs Qwen3.8 Max

> Qwen3.8 Max is the stronger model overall, scoring 56.0 to 38.6 on the Noometry Index. Mistral Large 3 costs 8.0× less per token, which makes it the better buy when Qwen3.8 Max's lead doesn't matter for your workload.

- Canonical page: https://noometry.com/compare/mistral-large-3-vs-qwen3-8-max
- Last updated: 2026-10-10
- Shared benchmarks: 20

## Summary

- They share 20 benchmarks with published results for both. Mistral Large 3 scores higher in 1 category and Qwen3.8 Max in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Qwen3.8 Max leads 53.4 to 14.6.
- The biggest single-benchmark swing is NYT Connections (extended): 7.5% for Mistral Large 3 and 88.3% for Qwen3.8 Max.
- Mistral Large 3 is cheaper at $0.25 / $0.75 per million input/output tokens, against $2 / $6 for Qwen3.8 Max.
- Qwen3.8 Max accepts more context: 1M tokens versus 262K.
- Mistral Large 3 has downloadable open weights; the other is API-only.

## Snapshot

| | Mistral Large 3 | Qwen3.8 Max |
|---|---|---|
| Provider | Mistral AI | Alibaba (Qwen) |
| Noometry Index | 38.6 | 56.0 |
| Rank | 172 | 21 |
| Context | 262K | 1M |
| Input $/M | $0.25 | $2 |
| Output $/M | $0.75 | $6 |
| Weights | Open | Proprietary |

## Coding

- Mistral Large 3: 34.1 (#230)
- Qwen3.8 Max: 52.3 (#30)

| Benchmark | Mistral Large 3 | Qwen3.8 Max |
|---|---|---|
| LMArena WebDev | 1230 | 1671 |
| LMArena Coding | 1447 | 1499 |
| DeepSWE | — | 57.5% |
| FrontierSWE | — | 15.8% |
| SciCode | — | 52.9% |

## Agentic & Tool Use

- Mistral Large 3: —
- Qwen3.8 Max: 42.6 (#17)

| Benchmark | Mistral Large 3 | Qwen3.8 Max |
|---|---|---|
| τ²-bench Banking | — | 55.1% |
| GDP.pdf | — | 23.2% |

## Reasoning

- Mistral Large 3: 14.6 (#314)
- Qwen3.8 Max: 53.4 (#26)

| Benchmark | Mistral Large 3 | Qwen3.8 Max |
|---|---|---|
| NYT Connections (extended) | 7.5% | 88.3% |
| LMArena Hard Prompts | 1429 | 1493 |
| Kagi LLM Benchmark | 50.9% | — |
| CritPt | — | 20% |
| Chess Puzzles | — | 40% |
| Thematic Generalization | 23% | — |
| Mystery Game Puzzles | — | 38% |
| DTBench | — | 92% |
| LMCA | — | 46.2% |
| Epoch Capabilities Index | — | 156.69 |

## Math

- Mistral Large 3: 36.0 (#131)
- Qwen3.8 Max: 73.1 (#15)

| Benchmark | Mistral Large 3 | Qwen3.8 Max |
|---|---|---|
| LMArena Math | 1415 | 1498 |
| FrontierMath (Tiers 1-3) | — | 74.7% |
| FrontierMath Tier 4 | — | 46.3% |
| OTIS Mock AIME 2024-2025 | — | 100% |
| ProofBench | — | 58% |

## Knowledge

- Mistral Large 3: 35.5 (#173)
- Qwen3.8 Max: 61.7 (#25)

| Benchmark | Mistral Large 3 | Qwen3.8 Max |
|---|---|---|
| LMArena Expert | 1421 | 1504 |
| GPQA Diamond | — | 92.7% |
| SimpleQA Verified | — | 47.3% |
| Vectara Hallucination Rate | 14.5% | — |

## Multimodal

- Mistral Large 3: 37.6 (#65)
- Qwen3.8 Max: 36.9 (#70)

| Benchmark | Mistral Large 3 | Qwen3.8 Max |
|---|---|---|
| LMArena Vision | 1222 | 1314 |
| Furniture Assembly | — | 20% |

## Multilingual

- Mistral Large 3: 52.6 (#79)
- Qwen3.8 Max: 56.6 (#18)

| Benchmark | Mistral Large 3 | Qwen3.8 Max |
|---|---|---|
| LMArena Non-English | 1412 | 1468 |
| LMArena Chinese | 1448 | 1537 |
| LMArena French | 1456 | 1502 |
| LMArena German | 1438 | 1489 |
| LMArena Japanese | 1399 | 1466 |
| LMArena Korean | 1384 | 1467 |
| LMArena Russian | 1412 | 1478 |
| LMArena Spanish | 1437 | 1495 |

## Instruction Following

- Mistral Large 3: 74.1 (#104)
- Qwen3.8 Max: 77.7 (#19)

| Benchmark | Mistral Large 3 | Qwen3.8 Max |
|---|---|---|
| LMArena Instruction Following | 1402 | 1478 |

## Long Context

- Mistral Large 3: 43.1 (#102)
- Qwen3.8 Max: 45.5 (#30)

| Benchmark | Mistral Large 3 | Qwen3.8 Max |
|---|---|---|
| LMArena Longer Query | 1413 | 1486 |

## Writing & Preference

- Mistral Large 3: 60.2 (#100)
- Qwen3.8 Max: 67.0 (#31)

| Benchmark | Mistral Large 3 | Qwen3.8 Max |
|---|---|---|
| LMArena Text | 1427 | 1480 |
| LMArena Creative Writing | 1385 | 1478 |
| LMArena Multi-Turn | 1429 | 1487 |
| EQ-Bench Creative Writing | 1412 | — |

## FAQ

### Is Mistral Large 3 better than Qwen3.8 Max?

Qwen3.8 Max is the stronger model overall, scoring 56.0 to 38.6 on the Noometry Index. Mistral Large 3 costs 8.0× less per token, which makes it the better buy when Qwen3.8 Max's lead doesn't matter for your workload.

### Which is cheaper, Mistral Large 3 or Qwen3.8 Max?

Mistral Large 3 is cheaper. It lists at $0.25 per million input tokens and $0.75 per million output tokens; Qwen3.8 Max lists at $2 and $6.

### Is Mistral Large 3 or Qwen3.8 Max better for coding?

Qwen3.8 Max scores higher on coding benchmarks: 52.3 versus 34.1 in the Noometry coding category.

### Which has the bigger context window?

Qwen3.8 Max does, with 1M tokens against 262K.

### How many benchmarks do Mistral Large 3 and Qwen3.8 Max share?

20 benchmarks have published results for both models. Mistral Large 3 has 24 scored results on Noometry and Qwen3.8 Max has 38.