Model comparison
Ministral 8B vs Qwen2.5-Coder-32B
Qwen2.5-Coder-32B is the stronger model overall, scoring 33.4 to 28.2 on the Noometry Index. Ministral 8B costs 5.0× less per token, which makes it the better buy when Qwen2.5-Coder-32B's lead doesn't matter for your workload.
Last verified . 12 shared benchmarks.
Mistral AI
Alibaba (Qwen)Summary
- They share 12 benchmarks with published results for both. Ministral 8B scores higher in 1 category and Qwen2.5-Coder-32B in 7 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Qwen2.5-Coder-32B leads 33.4 to 12.6.
- Ministral 8B is cheaper at $0.15 / $0.15 per million input/output tokens, against $0.66 / $1 for Qwen2.5-Coder-32B.
- Ministral 8B accepts more context: 262K tokens versus 33K.
Side by side
| Ministral 8B | Qwen2.5-Coder-32B | |
|---|---|---|
| Provider | Mistral AI | Alibaba (Qwen) |
| Noometry Index | 28.2 | 33.4 |
| Released | 2024-10-01 | 2024-09-18 |
| Weights | Open | Open |
| Context window | 262K | 33K |
| Max output | 262K | 29K |
| Input $ / M tokens | $0.15 | $0.66 |
| Output $ / M tokens | $0.15 | $1 |
| Results tracked | 17 | 31 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Ministral 8B leads
Ministral 8B: 35.0 (#230), Qwen2.5-Coder-32B: 22.6 (#333)
| Benchmark | Ministral 8B | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Coding | 1202 | 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% |
Agentic & Tool Use Not comparable
Ministral 8B: 16.4 (#148), Qwen2.5-Coder-32B: —
| Benchmark | Ministral 8B | Qwen2.5-Coder-32B |
|---|---|---|
| Berkeley Function Calling Leaderboard | 11.1% | — |
Reasoning Qwen2.5-Coder-32B leads
Ministral 8B: 18.4 (#281), Qwen2.5-Coder-32B: 21.2 (#225)
| Benchmark | Ministral 8B | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Hard Prompts | 1191 | 1251 |
| LiveBench Reasoning | — | 42.1% |
| DTBench | 45.7% | — |
| LiveBench Data Analysis | — | 49.9% |
| Epoch Capabilities Index | — | 119.49 |
| HellaSwag | — | 83% |
| LiveBench | — | 46.2% |
| WinoGrande | — | 80.8% |
Math Qwen2.5-Coder-32B leads
Ministral 8B: 25.7 (#267), Qwen2.5-Coder-32B: 33.3 (#204)
| Benchmark | Ministral 8B | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Math | 1188 | 1251 |
| LiveBench Math | — | 46.6% |
| MATH Level 5 | 14.9% | — |
| GSM8K | — | 93% |
Knowledge Qwen2.5-Coder-32B leads
Ministral 8B: 12.6 (#297), Qwen2.5-Coder-32B: 33.4 (#203)
| Benchmark | Ministral 8B | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Expert | 1170 | 1221 |
| GPQA Diamond | 27.1% | — |
| Vectara Hallucination Rate | 7.4% | — |
| ARC (AI2) Challenge | — | 70.5% |
| MMLU | — | 79.1% |
Multilingual Qwen2.5-Coder-32B leads
Ministral 8B: 35.1 (#247), Qwen2.5-Coder-32B: 37.8 (#235)
| Benchmark | Ministral 8B | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Non-English | 1165 | 1205 |
| LMArena Chinese | 1193 | 1222 |
| LMArena Russian | 1195 | 1228 |
Instruction Following Too close to call
Ministral 8B: 60.5 (#250), Qwen2.5-Coder-32B: 61.4 (#245)
| Benchmark | Ministral 8B | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Instruction Following | 1161 | 1223 |
| LiveBench Instruction Following | — | 58.7% |
Long Context Qwen2.5-Coder-32B leads
Ministral 8B: 36.7 (#227), Qwen2.5-Coder-32B: 38.0 (#208)
| Benchmark | Ministral 8B | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Longer Query | 1212 | 1251 |
Writing & Preference Qwen2.5-Coder-32B leads
Ministral 8B: 39.6 (#246), Qwen2.5-Coder-32B: 41.6 (#240)
| Benchmark | Ministral 8B | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Text | 1191 | 1230 |
| LMArena Creative Writing | 1175 | 1174 |
| LMArena Multi-Turn | 1166 | 1222 |
| LiveBench Language | — | 23.3% |
Frequently asked questions
Is Ministral 8B better than Qwen2.5-Coder-32B?
Qwen2.5-Coder-32B is the stronger model overall, scoring 33.4 to 28.2 on the Noometry Index. Ministral 8B costs 5.0× 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, Ministral 8B or Qwen2.5-Coder-32B?
Ministral 8B is cheaper. It lists at $0.15 per million input tokens and $0.15 per million output tokens; Qwen2.5-Coder-32B lists at $0.66 and $1.
Is Ministral 8B or Qwen2.5-Coder-32B better for coding?
Ministral 8B scores higher on coding benchmarks: 35.0 versus 22.6 in the Noometry coding category.
Which has the bigger context window?
Ministral 8B does, with 262K tokens against 33K.
How many benchmarks do Ministral 8B and Qwen2.5-Coder-32B share?
12 benchmarks have published results for both models. Ministral 8B has 17 scored results on Noometry and Qwen2.5-Coder-32B has 31.