Model comparison
Mistral 7B vs Qwen2.5-Coder-32B
Qwen2.5-Coder-32B is the stronger model overall, scoring 33.4 to 23.0 on the Noometry Index. Mistral 7B costs 3.0× less per token, which makes it the better buy when Qwen2.5-Coder-32B's lead doesn't matter for your workload.
Last verified . 22 shared benchmarks.
Mistral AI
Alibaba (Qwen)Summary
- They share 22 benchmarks with published results for both. Mistral 7B scores higher in 1 category and Qwen2.5-Coder-32B in 7 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Qwen2.5-Coder-32B leads 33.4 to 7.4.
- The biggest single-benchmark swing is BigCodeBench Complete: 27.3% for Mistral 7B and 58% for Qwen2.5-Coder-32B.
- Mistral 7B is cheaper at $0.25 / $0.25 per million input/output tokens, against $0.66 / $1 for Qwen2.5-Coder-32B.
- Qwen2.5-Coder-32B accepts more context: 33K tokens versus 8K.
Side by side
| Mistral 7B | Qwen2.5-Coder-32B | |
|---|---|---|
| Provider | Mistral AI | Alibaba (Qwen) |
| Noometry Index | 23.0 | 33.4 |
| Released | 2023-09-27 | 2024-09-18 |
| Weights | Open | Open |
| Context window | 8K | 33K |
| Max output | 8K | 29K |
| Input $ / M tokens | $0.25 | $0.66 |
| Output $ / M tokens | $0.25 | $1 |
| Results tracked | 37 | 31 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Mistral 7B leads
Mistral 7B: 26.4 (#326), Qwen2.5-Coder-32B: 22.6 (#333)
| Benchmark | Mistral 7B | Qwen2.5-Coder-32B |
|---|---|---|
| BigCodeBench Instruct | 19.5% | 49% |
| LMArena Coding | 1082 | 1276 |
| BigCodeBench Complete | 27.3% | 58% |
| HumanEval+ | 36% | 87.2% |
| MBPP+ | 42.1% | 77% |
| SWE-bench Verified (bash only) | — | 9% |
| Aider Polyglot | — | 16.4% |
| LiveBench Coding | — | 56.9% |
Reasoning Qwen2.5-Coder-32B leads
Mistral 7B: 13.1 (#336), Qwen2.5-Coder-32B: 21.2 (#225)
| Benchmark | Mistral 7B | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Hard Prompts | 1067 | 1251 |
| Epoch Capabilities Index | 112.21 | 119.49 |
| HellaSwag | 81% | 83% |
| WinoGrande | 75.3% | 80.8% |
| Chess Puzzles | 0% | — |
| LiveBench Reasoning | — | 42.1% |
| DTBench | 42.5% | — |
| LiveBench Data Analysis | — | 49.9% |
| Adversarial NLI | 47.1% | — |
| BIG-Bench Hard | 56.1% | — |
| LiveBench | — | 46.2% |
| PIQA | 83% | — |
Math Qwen2.5-Coder-32B leads
Mistral 7B: 8.1 (#325), Qwen2.5-Coder-32B: 33.3 (#204)
| Benchmark | Mistral 7B | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Math | 1085 | 1251 |
| GSM8K | 54.4% | 93% |
| OTIS Mock AIME 2024-2025 | 0.3% | — |
| LiveBench Math | — | 46.6% |
| MATH Level 5 | 3.7% | — |
Knowledge Qwen2.5-Coder-32B leads
Mistral 7B: 7.4 (#311), Qwen2.5-Coder-32B: 33.4 (#203)
| Benchmark | Mistral 7B | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Expert | 1036 | 1221 |
| ARC (AI2) Challenge | 78.6% | 70.5% |
| MMLU | 62.5% | 79.1% |
| GPQA Diamond | 15.2% | — |
| BoolQ | 87.4% | — |
| OpenBookQA | 79.8% | — |
| TriviaQA | 75.2% | — |
Multilingual Qwen2.5-Coder-32B leads
Mistral 7B: 25.8 (#283), Qwen2.5-Coder-32B: 37.8 (#235)
| Benchmark | Mistral 7B | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Non-English | 1012 | 1205 |
| LMArena Chinese | 1009 | 1222 |
| LMArena Russian | 1018 | 1228 |
| LMArena French | 1037 | — |
| LMArena German | 987 | — |
| LMArena Japanese | 878 | — |
| LMArena Spanish | 1026 | — |
Instruction Following Qwen2.5-Coder-32B leads
Mistral 7B: 54.2 (#280), Qwen2.5-Coder-32B: 61.4 (#245)
| Benchmark | Mistral 7B | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Instruction Following | 1060 | 1223 |
| LiveBench Instruction Following | — | 58.7% |
Long Context Qwen2.5-Coder-32B leads
Mistral 7B: 32.2 (#271), Qwen2.5-Coder-32B: 38.0 (#208)
| Benchmark | Mistral 7B | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Longer Query | 1060 | 1251 |
Writing & Preference Qwen2.5-Coder-32B leads
Mistral 7B: 30.7 (#286), Qwen2.5-Coder-32B: 41.6 (#240)
| Benchmark | Mistral 7B | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Text | 1090 | 1230 |
| LMArena Creative Writing | 1068 | 1174 |
| LMArena Multi-Turn | 1062 | 1222 |
| LiveBench Language | — | 23.3% |
Frequently asked questions
Is Mistral 7B better than Qwen2.5-Coder-32B?
Qwen2.5-Coder-32B is the stronger model overall, scoring 33.4 to 23.0 on the Noometry Index. Mistral 7B costs 3.0× 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 7B or Qwen2.5-Coder-32B?
Mistral 7B is cheaper. It lists at $0.25 per million input tokens and $0.25 per million output tokens; Qwen2.5-Coder-32B lists at $0.66 and $1.
Is Mistral 7B or Qwen2.5-Coder-32B better for coding?
Mistral 7B scores higher on coding benchmarks: 26.4 versus 22.6 in the Noometry coding category.
Which has the bigger context window?
Qwen2.5-Coder-32B does, with 33K tokens against 8K.
How many benchmarks do Mistral 7B and Qwen2.5-Coder-32B share?
22 benchmarks have published results for both models. Mistral 7B has 37 scored results on Noometry and Qwen2.5-Coder-32B has 31.