Model comparison
Mistral Medium vs Qwen2.5-Coder-32B
Mistral Medium is the stronger model overall, scoring 36.3 to 33.4 on the Noometry Index. Qwen2.5-Coder-32B costs 4.0× less per token, which makes it the better buy when Mistral Medium's lead doesn't matter for your workload.
Last verified . 12 shared benchmarks.
Mistral AI
Alibaba (Qwen)Summary
- They share 12 benchmarks with published results for both. Mistral Medium scores higher in 6 categories and Qwen2.5-Coder-32B in 2 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Mistral Medium leads 60.0 to 41.6.
- Qwen2.5-Coder-32B is cheaper at $0.66 / $1 per million input/output tokens, against $1.50 / $7.50 for Mistral Medium.
- Mistral Medium accepts more context: 262K tokens versus 33K.
Side by side
| Mistral Medium | Qwen2.5-Coder-32B | |
|---|---|---|
| Provider | Mistral AI | Alibaba (Qwen) |
| Noometry Index | 36.3 | 33.4 |
| Released | 2023-12-11 | 2024-09-18 |
| Weights | Open | Open |
| Context window | 262K | 33K |
| Max output | 262K | 29K |
| Input $ / M tokens | $1.50 | $0.66 |
| Output $ / M tokens | $7.50 | $1 |
| Results tracked | 36 | 31 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Mistral Medium leads
Mistral Medium: 34.2 (#243), Qwen2.5-Coder-32B: 22.6 (#333)
| Benchmark | Mistral Medium | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Coding | 1434 | 1276 |
| FrontierCode | 8% | — |
| SWE-bench Verified (bash only) | — | 9% |
| Aider Polyglot | — | 16.4% |
| SciCode | 40.2% | — |
| WeirdML | 43.7% | — |
| BigCodeBench Instruct | — | 49% |
| LiveBench Coding | — | 56.9% |
| BigCodeBench Complete | — | 58% |
| ALE-Bench | 763.98 | — |
| HumanEval+ | — | 87.2% |
| MBPP+ | — | 77% |
Agentic & Tool Use Not comparable
Mistral Medium: 28.3 (#90), Qwen2.5-Coder-32B: —
| Benchmark | Mistral Medium | Qwen2.5-Coder-32B |
|---|---|---|
| Berkeley Function Calling Leaderboard | 37.7% | — |
Reasoning Mistral Medium leads
Mistral Medium: 24.0 (#167), Qwen2.5-Coder-32B: 21.2 (#225)
| Benchmark | Mistral Medium | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Hard Prompts | 1426 | 1251 |
| Kagi LLM Benchmark | 50% | — |
| CritPt | 0% | — |
| LiveBench Reasoning | — | 42.1% |
| DTBench | 75.5% | — |
| LiveBench Data Analysis | — | 49.9% |
| LMCA | 26.1% | — |
| Surface Evolver Bench | 26.9% | — |
| Epoch Capabilities Index | — | 119.49 |
| HellaSwag | — | 83% |
| LiveBench | — | 46.2% |
| WinoGrande | — | 80.8% |
Math Qwen2.5-Coder-32B leads
Mistral Medium: 28.1 (#245), Qwen2.5-Coder-32B: 33.3 (#204)
| Benchmark | Mistral Medium | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Math | 1408 | 1251 |
| OTIS Mock AIME 2024-2025 | 32.2% | — |
| ProofBench | 9% | — |
| LiveBench Math | — | 46.6% |
| MATH Level 5 | 81.6% | — |
| FrontierMath (Feb 2025 set) | 0.3% | — |
| GSM8K | — | 93% |
Knowledge Qwen2.5-Coder-32B leads
Mistral Medium: 25.0 (#265), Qwen2.5-Coder-32B: 33.4 (#203)
| Benchmark | Mistral Medium | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Expert | 1408 | 1221 |
| GPQA Diamond | 59.5% | — |
| Humanity's Last Exam | 4.5% | — |
| Vectara Hallucination Rate | 22.7% | — |
| ARC (AI2) Challenge | — | 70.5% |
| MMLU | — | 79.1% |
Multimodal Not comparable
Mistral Medium: 35.3 (#88), Qwen2.5-Coder-32B: —
| Benchmark | Mistral Medium | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Vision | 1172 | — |
Multilingual Mistral Medium leads
Mistral Medium: 52.1 (#91), Qwen2.5-Coder-32B: 37.8 (#235)
| Benchmark | Mistral Medium | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Non-English | 1408 | 1205 |
| LMArena Chinese | 1447 | 1222 |
| LMArena Russian | 1411 | 1228 |
| LMArena French | 1459 | — |
| LMArena German | 1432 | — |
| LMArena Japanese | 1378 | — |
| LMArena Korean | 1380 | — |
| LMArena Spanish | 1433 | — |
Instruction Following Mistral Medium leads
Mistral Medium: 73.7 (#116), Qwen2.5-Coder-32B: 61.4 (#245)
| Benchmark | Mistral Medium | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Instruction Following | 1398 | 1223 |
| LiveBench Instruction Following | — | 58.7% |
Long Context Mistral Medium leads
Mistral Medium: 42.9 (#114), Qwen2.5-Coder-32B: 38.0 (#208)
| Benchmark | Mistral Medium | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Longer Query | 1406 | 1251 |
Writing & Preference Mistral Medium leads
Mistral Medium: 60.0 (#103), Qwen2.5-Coder-32B: 41.6 (#240)
| Benchmark | Mistral Medium | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Text | 1424 | 1230 |
| LMArena Creative Writing | 1391 | 1174 |
| LMArena Multi-Turn | 1418 | 1222 |
| Short-Story Creative Writing | 77.3% | — |
| LiveBench Language | — | 23.3% |
Frequently asked questions
Is Mistral Medium better than Qwen2.5-Coder-32B?
Mistral Medium is the stronger model overall, scoring 36.3 to 33.4 on the Noometry Index. Qwen2.5-Coder-32B costs 4.0× less per token, which makes it the better buy when Mistral Medium's lead doesn't matter for your workload.
Which is cheaper, Mistral Medium or Qwen2.5-Coder-32B?
Qwen2.5-Coder-32B is cheaper. It lists at $0.66 per million input tokens and $1 per million output tokens; Mistral Medium lists at $1.50 and $7.50.
Is Mistral Medium or Qwen2.5-Coder-32B better for coding?
Mistral Medium scores higher on coding benchmarks: 34.2 versus 22.6 in the Noometry coding category.
Which has the bigger context window?
Mistral Medium does, with 262K tokens against 33K.
How many benchmarks do Mistral Medium and Qwen2.5-Coder-32B share?
12 benchmarks have published results for both models. Mistral Medium has 36 scored results on Noometry and Qwen2.5-Coder-32B has 31.