Model comparison
Mistral Medium vs Qwen3.8 Max
Qwen3.8 Max is the stronger model overall, scoring 56.8 to 36.3 on the Noometry Index.
Last verified . 25 shared benchmarks.
Mistral AI
Alibaba (Qwen)Summary
- They share 25 benchmarks with published results for both. Mistral Medium scores higher in 0 categories and Qwen3.8 Max in 10 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in math, where Qwen3.8 Max leads 73.2 to 28.1.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 32.2% for Mistral Medium and 100% for Qwen3.8 Max.
- Both cost about the same: $1.50 input and $7.50 output per million tokens.
- Qwen3.8 Max accepts more context: 1M tokens versus 262K.
- Mistral Medium has downloadable open weights; the other is API-only.
Side by side
| Mistral Medium | Qwen3.8 Max | |
|---|---|---|
| Provider | Mistral AI | Alibaba (Qwen) |
| Noometry Index | 36.3 | 56.8 |
| Released | 2023-12-11 | 2026-08-02 |
| Weights | Open | Proprietary |
| Context window | 262K | 1M |
| Max output | 262K | 131K |
| Input $ / M tokens | $1.50 | $2 |
| Output $ / M tokens | $7.50 | $6 |
| Results tracked | 36 | 39 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen3.8 Max leads
Mistral Medium: 34.2 (#243), Qwen3.8 Max: 53.5 (#29)
| Benchmark | Mistral Medium | Qwen3.8 Max |
|---|---|---|
| SciCode | 40.2% | 53.2% |
| LMArena Coding | 1434 | 1502 |
| DeepSWE | — | 57.5% |
| FrontierCode | 8% | — |
| LMArena WebDev | — | 1674 |
| FrontierSWE | — | 17.8% |
| WeirdML | 43.7% | — |
| ALE-Bench | 763.98 | — |
Agentic & Tool Use Qwen3.8 Max leads
Mistral Medium: 28.3 (#90), Qwen3.8 Max: 45.4 (#14)
| Benchmark | Mistral Medium | Qwen3.8 Max |
|---|---|---|
| APEX-Agents | — | 63.3% |
| Berkeley Function Calling Leaderboard | 37.7% | — |
| τ²-bench Banking | — | 55.1% |
| GDP.pdf | — | 23.2% |
Reasoning Qwen3.8 Max leads
Mistral Medium: 24.0 (#167), Qwen3.8 Max: 54.4 (#26)
| Benchmark | Mistral Medium | Qwen3.8 Max |
|---|---|---|
| CritPt | 0% | 20% |
| LMArena Hard Prompts | 1426 | 1496 |
| DTBench | 75.5% | 92% |
| LMCA | 26.1% | 46.2% |
| Kagi LLM Benchmark | 50% | — |
| NYT Connections (extended) | — | 88.3% |
| Chess Puzzles | — | 40% |
| Mystery Game Puzzles | — | 38% |
| Surface Evolver Bench | 26.9% | — |
| Epoch Capabilities Index | — | 156.41 |
Math Qwen3.8 Max leads
Mistral Medium: 28.1 (#245), Qwen3.8 Max: 73.2 (#20)
| Benchmark | Mistral Medium | Qwen3.8 Max |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 32.2% | 100% |
| ProofBench | 9% | 58% |
| LMArena Math | 1408 | 1499 |
| FrontierMath (Tiers 1-3) | — | 74.7% |
| FrontierMath Tier 4 | — | 46.3% |
| MATH Level 5 | 81.6% | — |
| FrontierMath (Feb 2025 set) | 0.3% | — |
Knowledge Qwen3.8 Max leads
Mistral Medium: 25.0 (#265), Qwen3.8 Max: 61.7 (#27)
| Benchmark | Mistral Medium | Qwen3.8 Max |
|---|---|---|
| GPQA Diamond | 59.5% | 92.7% |
| LMArena Expert | 1408 | 1507 |
| Humanity's Last Exam | 4.5% | — |
| SimpleQA Verified | — | 47.3% |
| Vectara Hallucination Rate | 22.7% | — |
Multimodal Qwen3.8 Max leads
Mistral Medium: 35.3 (#88), Qwen3.8 Max: 37.2 (#75)
| Benchmark | Mistral Medium | Qwen3.8 Max |
|---|---|---|
| LMArena Vision | 1172 | 1314 |
| Furniture Assembly | — | 20% |
Multilingual Qwen3.8 Max leads
Mistral Medium: 52.1 (#91), Qwen3.8 Max: 56.7 (#18)
| Benchmark | Mistral Medium | Qwen3.8 Max |
|---|---|---|
| LMArena Non-English | 1408 | 1472 |
| LMArena Chinese | 1447 | 1538 |
| LMArena French | 1459 | 1503 |
| LMArena German | 1432 | 1483 |
| LMArena Japanese | 1378 | 1467 |
| LMArena Korean | 1380 | 1461 |
| LMArena Russian | 1411 | 1481 |
| LMArena Spanish | 1433 | 1492 |
Instruction Following Qwen3.8 Max leads
Mistral Medium: 73.7 (#116), Qwen3.8 Max: 77.6 (#17)
| Benchmark | Mistral Medium | Qwen3.8 Max |
|---|---|---|
| LMArena Instruction Following | 1398 | 1479 |
Long Context Qwen3.8 Max leads
Mistral Medium: 42.9 (#114), Qwen3.8 Max: 45.6 (#31)
| Benchmark | Mistral Medium | Qwen3.8 Max |
|---|---|---|
| LMArena Longer Query | 1406 | 1489 |
Writing & Preference Qwen3.8 Max leads
Mistral Medium: 60.0 (#103), Qwen3.8 Max: 67.1 (#30)
| Benchmark | Mistral Medium | Qwen3.8 Max |
|---|---|---|
| LMArena Text | 1424 | 1483 |
| LMArena Creative Writing | 1391 | 1479 |
| LMArena Multi-Turn | 1418 | 1489 |
| Short-Story Creative Writing | 77.3% | — |
Frequently asked questions
Is Mistral Medium better than Qwen3.8 Max?
Qwen3.8 Max is the stronger model overall, scoring 56.8 to 36.3 on the Noometry Index.
Which is cheaper, Mistral Medium or Qwen3.8 Max?
Qwen3.8 Max is cheaper. It lists at $2 per million input tokens and $6 per million output tokens; Mistral Medium lists at $1.50 and $7.50.
Is Mistral Medium or Qwen3.8 Max better for coding?
Qwen3.8 Max scores higher on coding benchmarks: 53.5 versus 34.2 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 Medium and Qwen3.8 Max share?
25 benchmarks have published results for both models. Mistral Medium has 36 scored results on Noometry and Qwen3.8 Max has 39.