Model comparison
Mistral Medium vs Qwen3.8 27B
Qwen3.8 27B is the stronger model overall, scoring 46.0 to 36.3 on the Noometry Index.
Last verified . 24 shared benchmarks.
Mistral AI
Alibaba (Qwen)Summary
- They share 24 benchmarks with published results for both. Mistral Medium scores higher in 0 categories and Qwen3.8 27B in 10 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Qwen3.8 27B leads 41.0 to 24.0.
- The biggest single-benchmark swing is Surface Evolver Bench: 26.9% for Mistral Medium and 45% for Qwen3.8 27B.
- Qwen3.8 27B is cheaper at $0.99 / $1.49 per million input/output tokens, against $1.50 / $7.50 for Mistral Medium.
Side by side
| Mistral Medium | Qwen3.8 27B | |
|---|---|---|
| Provider | Mistral AI | Alibaba (Qwen) |
| Noometry Index | 36.3 | 46.0 |
| Released | 2023-12-11 | 2026-08-14 |
| Weights | Open | Open |
| Context window | 262K | 262K |
| Max output | 262K | 33K |
| Input $ / M tokens | $1.50 | $0.99 |
| Output $ / M tokens | $7.50 | $1.49 |
| Results tracked | 36 | 31 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen3.8 27B leads
Mistral Medium: 34.2 (#243), Qwen3.8 27B: 50.5 (#44)
| Benchmark | Mistral Medium | Qwen3.8 27B |
|---|---|---|
| SciCode | 40.2% | 46.6% |
| LMArena Coding | 1434 | 1482 |
| FrontierCode | 8% | — |
| LMArena WebDev | — | 1593 |
| WeirdML | 43.7% | — |
| ALE-Bench | 763.98 | — |
Agentic & Tool Use Qwen3.8 27B leads
Mistral Medium: 28.3 (#90), Qwen3.8 27B: 32.9 (#57)
| Benchmark | Mistral Medium | Qwen3.8 27B |
|---|---|---|
| APEX-Agents | — | 47.5% |
| Berkeley Function Calling Leaderboard | 37.7% | — |
Reasoning Qwen3.8 27B leads
Mistral Medium: 24.0 (#167), Qwen3.8 27B: 41.0 (#54)
| Benchmark | Mistral Medium | Qwen3.8 27B |
|---|---|---|
| CritPt | 0% | 5.4% |
| LMArena Hard Prompts | 1426 | 1460 |
| DTBench | 75.5% | 88% |
| LMCA | 26.1% | 41.4% |
| Surface Evolver Bench | 26.9% | 45% |
| ARC-AGI-2 | — | 42.4% |
| Kagi LLM Benchmark | 50% | — |
| NYT Connections (extended) | — | 54.5% |
| ARC-AGI-1 | — | 87.5% |
| Epoch Capabilities Index | — | 149.38 |
Math Qwen3.8 27B leads
Mistral Medium: 28.1 (#245), Qwen3.8 27B: 37.1 (#161)
| Benchmark | Mistral Medium | Qwen3.8 27B |
|---|---|---|
| ProofBench | 9% | 16% |
| LMArena Math | 1408 | 1456 |
| OTIS Mock AIME 2024-2025 | 32.2% | — |
| MATH Level 5 | 81.6% | — |
| FrontierMath (Feb 2025 set) | 0.3% | — |
Knowledge Qwen3.8 27B leads
Mistral Medium: 25.0 (#265), Qwen3.8 27B: 41.6 (#109)
| Benchmark | Mistral Medium | Qwen3.8 27B |
|---|---|---|
| LMArena Expert | 1408 | 1482 |
| GPQA Diamond | 59.5% | — |
| Humanity's Last Exam | 4.5% | — |
| Vectara Hallucination Rate | 22.7% | — |
Multimodal Qwen3.8 27B leads
Mistral Medium: 35.3 (#88), Qwen3.8 27B: 41.3 (#37)
| Benchmark | Mistral Medium | Qwen3.8 27B |
|---|---|---|
| LMArena Vision | 1172 | 1271 |
Multilingual Qwen3.8 27B leads
Mistral Medium: 52.1 (#91), Qwen3.8 27B: 53.7 (#60)
| Benchmark | Mistral Medium | Qwen3.8 27B |
|---|---|---|
| LMArena Non-English | 1408 | 1430 |
| LMArena Chinese | 1447 | 1504 |
| LMArena French | 1459 | 1465 |
| LMArena German | 1432 | 1438 |
| LMArena Japanese | 1378 | 1384 |
| LMArena Korean | 1380 | 1393 |
| LMArena Russian | 1411 | 1415 |
| LMArena Spanish | 1433 | 1448 |
Instruction Following Qwen3.8 27B leads
Mistral Medium: 73.7 (#116), Qwen3.8 27B: 75.8 (#53)
| Benchmark | Mistral Medium | Qwen3.8 27B |
|---|---|---|
| LMArena Instruction Following | 1398 | 1439 |
Long Context Qwen3.8 27B leads
Mistral Medium: 42.9 (#114), Qwen3.8 27B: 44.3 (#70)
| Benchmark | Mistral Medium | Qwen3.8 27B |
|---|---|---|
| LMArena Longer Query | 1406 | 1450 |
Writing & Preference Qwen3.8 27B leads
Mistral Medium: 60.0 (#103), Qwen3.8 27B: 65.8 (#43)
| Benchmark | Mistral Medium | Qwen3.8 27B |
|---|---|---|
| LMArena Text | 1424 | 1441 |
| LMArena Creative Writing | 1391 | 1384 |
| LMArena Multi-Turn | 1418 | 1441 |
| Short-Story Creative Writing | 77.3% | — |
| EQ-Bench Creative Writing | — | 1671 |
Frequently asked questions
Is Mistral Medium better than Qwen3.8 27B?
Qwen3.8 27B is the stronger model overall, scoring 46.0 to 36.3 on the Noometry Index.
Which is cheaper, Mistral Medium or Qwen3.8 27B?
Qwen3.8 27B is cheaper. It lists at $0.99 per million input tokens and $1.49 per million output tokens; Mistral Medium lists at $1.50 and $7.50.
Is Mistral Medium or Qwen3.8 27B better for coding?
Qwen3.8 27B scores higher on coding benchmarks: 50.5 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 Qwen3.8 27B share?
24 benchmarks have published results for both models. Mistral Medium has 36 scored results on Noometry and Qwen3.8 27B has 31.