Model comparison
Mistral Small 3.1 vs Qwen2.5-Coder-32B
Qwen2.5-Coder-32B is the stronger model overall, scoring 32.9 to 31.4 on the Noometry Index. Mistral Small 3.1 costs 1.9× less per token, which makes it the better buy when Qwen2.5-Coder-32B's lead doesn't matter for your workload.
Last verified . 13 shared benchmarks.
Mistral AI
Alibaba (Qwen)Summary
- They share 13 benchmarks with published results for both. Mistral Small 3.1 scores higher in 4 categories and Qwen2.5-Coder-32B in 4 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Qwen2.5-Coder-32B leads 30.8 to 13.7.
- Mistral Small 3.1 is cheaper at $0.35 / $0.56 per million input/output tokens, against $0.66 / $1 for Qwen2.5-Coder-32B.
- Mistral Small 3.1 accepts more context: 128K tokens versus 33K.
Side by side
| Mistral Small 3.1 | Qwen2.5-Coder-32B | |
|---|---|---|
| Provider | Mistral AI | Alibaba (Qwen) |
| Noometry Index | 31.4 | 32.9 |
| Released | 2025-03-17 | 2024-09-18 |
| Weights | Open | Open |
| Context window | 128K | 33K |
| Max output | 102K | 29K |
| Input $ / M tokens | $0.35 | $0.66 |
| Output $ / M tokens | $0.56 | $1 |
| Results tracked | 28 | 31 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Mistral Small 3.1 leads
Mistral Small 3.1: 37.7 (#175), Qwen2.5-Coder-32B: 22.4 (#328)
| Benchmark | Mistral Small 3.1 | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Coding | 1309 | 1276 |
| SWE-bench Verified (bash only) | — | 9% |
| Aider Polyglot | — | 16.4% |
| BigCodeBench Instruct | — | 49% |
| LiveBench Coding | — | 56.9% |
| BigCodeBench Complete | — | 58% |
| HumanEval+ | — | 87.2% |
| MBPP+ | — | 77% |
Reasoning Qwen2.5-Coder-32B leads
Mistral Small 3.1: 18.9 (#249), Qwen2.5-Coder-32B: 20.4 (#220)
| Benchmark | Mistral Small 3.1 | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Hard Prompts | 1278 | 1251 |
| Epoch Capabilities Index | 127.48 | 119.41 |
| Chess Puzzles | 1% | — |
| LiveBench Reasoning | — | 42.1% |
| LiveBench Data Analysis | — | 49.9% |
| HellaSwag | — | 83% |
| LiveBench | — | 46.2% |
| WinoGrande | — | 80.8% |
Math Qwen2.5-Coder-32B leads
Mistral Small 3.1: 13.7 (#295), Qwen2.5-Coder-32B: 30.8 (#203)
| Benchmark | Mistral Small 3.1 | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Math | 1261 | 1251 |
| OTIS Mock AIME 2024-2025 | 3.9% | — |
| Omni-MATH | 24.8% | — |
| LiveBench Math | — | 46.6% |
| GSM8K | — | 93% |
Knowledge Qwen2.5-Coder-32B leads
Mistral Small 3.1: 22.6 (#266), Qwen2.5-Coder-32B: 32.9 (#200)
| Benchmark | Mistral Small 3.1 | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Expert | 1258 | 1221 |
| GPQA Diamond | 41.9% | — |
| MMLU-Pro | 61% | — |
| GPQA (HELM) | 39.2% | — |
| ARC (AI2) Challenge | — | 70.5% |
| MMLU | — | 79.1% |
Multimodal Not comparable
Mistral Small 3.1: 32.6 (#93), Qwen2.5-Coder-32B: —
| Benchmark | Mistral Small 3.1 | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Vision | 1136 | — |
Multilingual Mistral Small 3.1 leads
Mistral Small 3.1: 41.4 (#204), Qwen2.5-Coder-32B: 38.0 (#230)
| Benchmark | Mistral Small 3.1 | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Non-English | 1255 | 1205 |
| LMArena Chinese | 1254 | 1222 |
| LMArena Russian | 1263 | 1228 |
| LMArena French | 1270 | — |
| LMArena German | 1266 | — |
| LMArena Japanese | 1210 | — |
| LMArena Korean | 1206 | — |
| LMArena Spanish | 1283 | — |
Instruction Following Mistral Small 3.1 leads
Mistral Small 3.1: 63.8 (#225), Qwen2.5-Coder-32B: 61.6 (#240)
| Benchmark | Mistral Small 3.1 | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Instruction Following | 1264 | 1223 |
| LiveBench Instruction Following | — | 58.7% |
| IFEval | 75% | — |
Long Context Mistral Small 3.1 leads
Mistral Small 3.1: 39.5 (#172), Qwen2.5-Coder-32B: 37.9 (#204)
| Benchmark | Mistral Small 3.1 | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Longer Query | 1300 | 1251 |
Writing & Preference Qwen2.5-Coder-32B leads
Mistral Small 3.1: 37.2 (#254), Qwen2.5-Coder-32B: 41.8 (#235)
| Benchmark | Mistral Small 3.1 | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Text | 1278 | 1230 |
| LMArena Creative Writing | 1253 | 1174 |
| LMArena Multi-Turn | 1270 | 1222 |
| EQ-Bench Creative Writing | 761 | — |
| WildBench | 78.8% | — |
| LiveBench Language | — | 23.3% |
Frequently asked questions
Is Mistral Small 3.1 better than Qwen2.5-Coder-32B?
Qwen2.5-Coder-32B is the stronger model overall, scoring 32.9 to 31.4 on the Noometry Index. Mistral Small 3.1 costs 1.9× less per token, which makes it the better buy when Qwen2.5-Coder-32B's lead doesn't matter for your workload.
Which is cheaper, Mistral Small 3.1 or Qwen2.5-Coder-32B?
Mistral Small 3.1 is cheaper. It lists at $0.35 per million input tokens and $0.56 per million output tokens; Qwen2.5-Coder-32B lists at $0.66 and $1.
Is Mistral Small 3.1 or Qwen2.5-Coder-32B better for coding?
Mistral Small 3.1 scores higher on coding benchmarks: 37.7 versus 22.4 in the Noometry coding category.
Which has the bigger context window?
Mistral Small 3.1 does, with 128K tokens against 33K.
How many benchmarks do Mistral Small 3.1 and Qwen2.5-Coder-32B share?
13 benchmarks have published results for both models. Mistral Small 3.1 has 28 scored results on Noometry and Qwen2.5-Coder-32B has 31.