Model comparison
Magistral Medium vs Qwen3.8 Max
Qwen3.8 Max is the stronger model overall, scoring 56.8 to 35.2 on the Noometry Index.
Last verified . 19 shared benchmarks.
Mistral AI
Alibaba (Qwen)Summary
- They share 19 benchmarks with published results for both. Magistral Medium scores higher in 0 categories 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 54.4 to 8.6.
- The biggest single-benchmark swing is CritPt: 0.3% for Magistral Medium and 20% for Qwen3.8 Max.
- Magistral Medium is cheaper at $2 / $5 per million input/output tokens, against $2 / $6 for Qwen3.8 Max.
- Qwen3.8 Max accepts more context: 1M tokens versus 262K.
- Magistral Medium has downloadable open weights; the other is API-only.
Side by side
| Magistral Medium | Qwen3.8 Max | |
|---|---|---|
| Provider | Mistral AI | Alibaba (Qwen) |
| Noometry Index | 35.2 | 56.8 |
| Released | 2025-03-17 | 2026-08-02 |
| Weights | Open | Proprietary |
| Context window | 262K | 1M |
| Max output | 16K | 131K |
| Input $ / M tokens | $2 | $2 |
| Output $ / M tokens | $5 | $6 |
| Results tracked | 22 | 39 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen3.8 Max leads
Magistral Medium: 39.1 (#161), Qwen3.8 Max: 53.5 (#29)
| Benchmark | Magistral Medium | Qwen3.8 Max |
|---|---|---|
| SciCode | 39.2% | 53.2% |
| LMArena Coding | 1319 | 1502 |
| DeepSWE | — | 57.5% |
| LMArena WebDev | — | 1674 |
| FrontierSWE | — | 17.8% |
Agentic & Tool Use Not comparable
Magistral Medium: —, Qwen3.8 Max: 45.4 (#14)
| Benchmark | Magistral Medium | Qwen3.8 Max |
|---|---|---|
| APEX-Agents | — | 63.3% |
| τ²-bench Banking | — | 55.1% |
| GDP.pdf | — | 23.2% |
Reasoning Qwen3.8 Max leads
Magistral Medium: 8.6 (#348), Qwen3.8 Max: 54.4 (#26)
| Benchmark | Magistral Medium | Qwen3.8 Max |
|---|---|---|
| CritPt | 0.3% | 20% |
| LMArena Hard Prompts | 1267 | 1496 |
| ARC-AGI-2 | 0% | — |
| Kagi LLM Benchmark | 16.2% | — |
| NYT Connections (extended) | — | 88.3% |
| ARC-AGI-1 | 6.1% | — |
| Chess Puzzles | — | 40% |
| Mystery Game Puzzles | — | 38% |
| DTBench | — | 92% |
| LMCA | — | 46.2% |
| Epoch Capabilities Index | — | 156.41 |
Math Qwen3.8 Max leads
Magistral Medium: 35.1 (#189), Qwen3.8 Max: 73.2 (#20)
| Benchmark | Magistral Medium | Qwen3.8 Max |
|---|---|---|
| LMArena Math | 1250 | 1499 |
| FrontierMath (Tiers 1-3) | — | 74.7% |
| FrontierMath Tier 4 | — | 46.3% |
| OTIS Mock AIME 2024-2025 | — | 100% |
| ProofBench | — | 58% |
Knowledge Qwen3.8 Max leads
Magistral Medium: 33.5 (#202), Qwen3.8 Max: 61.7 (#27)
| Benchmark | Magistral Medium | Qwen3.8 Max |
|---|---|---|
| LMArena Expert | 1223 | 1507 |
| GPQA Diamond | — | 92.7% |
| SimpleQA Verified | — | 47.3% |
Multimodal Not comparable
Magistral Medium: —, Qwen3.8 Max: 37.2 (#75)
| Benchmark | Magistral Medium | Qwen3.8 Max |
|---|---|---|
| LMArena Vision | — | 1314 |
| Furniture Assembly | — | 20% |
Multilingual Qwen3.8 Max leads
Magistral Medium: 39.6 (#224), Qwen3.8 Max: 56.7 (#18)
| Benchmark | Magistral Medium | Qwen3.8 Max |
|---|---|---|
| LMArena Non-English | 1232 | 1472 |
| LMArena Chinese | 1227 | 1538 |
| LMArena French | 1267 | 1503 |
| LMArena German | 1248 | 1483 |
| LMArena Japanese | 1175 | 1467 |
| LMArena Korean | 1125 | 1461 |
| LMArena Russian | 1224 | 1481 |
| LMArena Spanish | 1271 | 1492 |
Instruction Following Qwen3.8 Max leads
Magistral Medium: 66.0 (#211), Qwen3.8 Max: 77.6 (#17)
| Benchmark | Magistral Medium | Qwen3.8 Max |
|---|---|---|
| LMArena Instruction Following | 1254 | 1479 |
Long Context Qwen3.8 Max leads
Magistral Medium: 39.3 (#183), Qwen3.8 Max: 45.6 (#31)
| Benchmark | Magistral Medium | Qwen3.8 Max |
|---|---|---|
| LMArena Longer Query | 1295 | 1489 |
Writing & Preference Qwen3.8 Max leads
Magistral Medium: 46.3 (#219), Qwen3.8 Max: 67.1 (#30)
| Benchmark | Magistral Medium | Qwen3.8 Max |
|---|---|---|
| LMArena Text | 1255 | 1483 |
| LMArena Creative Writing | 1245 | 1479 |
| LMArena Multi-Turn | 1275 | 1489 |
Frequently asked questions
Is Magistral Medium better than Qwen3.8 Max?
Qwen3.8 Max is the stronger model overall, scoring 56.8 to 35.2 on the Noometry Index.
Which is cheaper, Magistral Medium or Qwen3.8 Max?
Magistral Medium is cheaper. It lists at $2 per million input tokens and $5 per million output tokens; Qwen3.8 Max lists at $2 and $6.
Is Magistral Medium or Qwen3.8 Max better for coding?
Qwen3.8 Max scores higher on coding benchmarks: 53.5 versus 39.1 in the Noometry coding category.
Which has the bigger context window?
Qwen3.8 Max does, with 1M tokens against 262K.
How many benchmarks do Magistral Medium and Qwen3.8 Max share?
19 benchmarks have published results for both models. Magistral Medium has 22 scored results on Noometry and Qwen3.8 Max has 39.