Model comparison
Mistral Medium vs Qwen3.5 122B-A10B
Qwen3.5 122B-A10B is the stronger model overall, scoring 41.6 to 36.0 on the Noometry Index.
Last verified . 23 shared benchmarks.
Mistral AI
Alibaba (Qwen)Summary
- They share 23 benchmarks with published results for both. Mistral Medium scores higher in 1 category and Qwen3.5 122B-A10B in 8 categories; 5 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Qwen3.5 122B-A10B leads 38.2 to 24.9.
- The biggest single-benchmark swing is Vectara Hallucination Rate: 22.7% for Mistral Medium and 11.2% for Qwen3.5 122B-A10B.
- Qwen3.5 122B-A10B is cheaper at $0.40 / $3.20 per million input/output tokens, against $1.50 / $7.50 for Mistral Medium.
Side by side
| Mistral Medium | Qwen3.5 122B-A10B | |
|---|---|---|
| Provider | Mistral AI | Alibaba (Qwen) |
| Noometry Index | 36.0 | 41.6 |
| Released | 2023-12-11 | 2026-02-23 |
| Weights | Open | Open |
| Context window | 262K | 262K |
| Max output | 262K | 66K |
| Input $ / M tokens | $1.50 | $0.40 |
| Output $ / M tokens | $7.50 | $3.20 |
| Results tracked | 36 | 27 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen3.5 122B-A10B leads
Mistral Medium: 33.8 (#241), Qwen3.5 122B-A10B: 38.8 (#155)
| Benchmark | Mistral Medium | Qwen3.5 122B-A10B |
|---|---|---|
| SciCode | 40.2% | 35.6% |
| LMArena Coding | 1434 | 1436 |
| FrontierCode | 8% | — |
| LMArena WebDev | — | 1360 |
| WeirdML | 43.7% | — |
| ALE-Bench | 763.98 | — |
Agentic & Tool Use Not comparable
Mistral Medium: 28.1 (#84), Qwen3.5 122B-A10B: —
| Benchmark | Mistral Medium | Qwen3.5 122B-A10B |
|---|---|---|
| Berkeley Function Calling Leaderboard | 37.7% | — |
Reasoning Qwen3.5 122B-A10B leads
Mistral Medium: 23.3 (#165), Qwen3.5 122B-A10B: 26.4 (#121)
| Benchmark | Mistral Medium | Qwen3.5 122B-A10B |
|---|---|---|
| CritPt | 0% | 0.9% |
| LMArena Hard Prompts | 1426 | 1421 |
| DTBench | 75.5% | 84.3% |
| LMCA | 26.1% | 32.2% |
| Kagi LLM Benchmark | 50% | — |
| NYT Connections (extended) | — | 51.7% |
| Thematic Generalization | — | 51.2% |
| Mystery Game Puzzles | — | 17% |
| Surface Evolver Bench | 26.9% | — |
Math Qwen3.5 122B-A10B leads
Mistral Medium: 27.2 (#238), Qwen3.5 122B-A10B: 36.4 (#114)
| Benchmark | Mistral Medium | Qwen3.5 122B-A10B |
|---|---|---|
| LMArena Math | 1409 | 1432 |
| OTIS Mock AIME 2024-2025 | 32.2% | — |
| ProofBench | 9% | — |
| MATH Level 5 | 81.6% | — |
| FrontierMath (Feb 2025 set) | 0.3% | — |
Knowledge Qwen3.5 122B-A10B leads
Mistral Medium: 24.9 (#260), Qwen3.5 122B-A10B: 38.2 (#138)
| Benchmark | Mistral Medium | Qwen3.5 122B-A10B |
|---|---|---|
| Vectara Hallucination Rate | 22.7% | 11.2% |
| LMArena Expert | 1408 | 1431 |
| GPQA Diamond | 59.5% | — |
| Humanity's Last Exam | 4.5% | — |
Multimodal Qwen3.5 122B-A10B leads
Mistral Medium: 34.6 (#82), Qwen3.5 122B-A10B: 39.0 (#52)
| Benchmark | Mistral Medium | Qwen3.5 122B-A10B |
|---|---|---|
| LMArena Vision | 1172 | 1245 |
Multilingual Too close to call
Mistral Medium: 52.3 (#86), Qwen3.5 122B-A10B: 51.8 (#101)
| Benchmark | Mistral Medium | Qwen3.5 122B-A10B |
|---|---|---|
| LMArena Non-English | 1408 | 1400 |
| LMArena Chinese | 1448 | 1462 |
| LMArena French | 1460 | 1442 |
| LMArena German | 1433 | 1425 |
| LMArena Japanese | 1378 | 1364 |
| LMArena Korean | 1382 | 1353 |
| LMArena Russian | 1411 | 1400 |
| LMArena Spanish | 1433 | 1423 |
Instruction Following Too close to call
Mistral Medium: 73.9 (#111), Qwen3.5 122B-A10B: 74.0 (#109)
| Benchmark | Mistral Medium | Qwen3.5 122B-A10B |
|---|---|---|
| LMArena Instruction Following | 1398 | 1399 |
Long Context Too close to call
Mistral Medium: 42.9 (#108), Qwen3.5 122B-A10B: 43.0 (#105)
| Benchmark | Mistral Medium | Qwen3.5 122B-A10B |
|---|---|---|
| LMArena Longer Query | 1406 | 1410 |
Writing & Preference Too close to call
Mistral Medium: 60.2 (#97), Qwen3.5 122B-A10B: 60.2 (#98)
| Benchmark | Mistral Medium | Qwen3.5 122B-A10B |
|---|---|---|
| LMArena Text | 1424 | 1417 |
| LMArena Creative Writing | 1391 | 1369 |
| LMArena Multi-Turn | 1418 | 1416 |
| Short-Story Creative Writing | 77.3% | — |
Frequently asked questions
Is Mistral Medium better than Qwen3.5 122B-A10B?
Qwen3.5 122B-A10B is the stronger model overall, scoring 41.6 to 36.0 on the Noometry Index.
Which is cheaper, Mistral Medium or Qwen3.5 122B-A10B?
Qwen3.5 122B-A10B is cheaper. It lists at $0.40 per million input tokens and $3.20 per million output tokens; Mistral Medium lists at $1.50 and $7.50.
Is Mistral Medium or Qwen3.5 122B-A10B better for coding?
Qwen3.5 122B-A10B scores higher on coding benchmarks: 38.8 versus 33.8 in the Noometry coding category.
Which has the bigger context window?
Both accept 262K tokens.
How many benchmarks do Mistral Medium and Qwen3.5 122B-A10B share?
23 benchmarks have published results for both models. Mistral Medium has 36 scored results on Noometry and Qwen3.5 122B-A10B has 27.