Model comparison
Ministral 3B vs Qwen3.8 Max
Qwen3.8 Max is the stronger model overall, scoring 56.8 to 26.2 on the Noometry Index. Ministral 3B costs 30× less per token, which makes it the better buy when Qwen3.8 Max's lead doesn't matter for your workload.
Last verified . 4 shared benchmarks.
Mistral AI
Alibaba (Qwen)Summary
- They share 4 benchmarks with published results for both. Ministral 3B scores higher in 0 categories and Qwen3.8 Max in 3 categories; 3 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Qwen3.8 Max leads 61.7 to 10.4.
- The biggest single-benchmark swing is GPQA Diamond: 25.3% for Ministral 3B and 92.7% for Qwen3.8 Max.
- Ministral 3B is cheaper at $0.10 / $0.10 per million input/output tokens, against $2 / $6 for Qwen3.8 Max.
- Qwen3.8 Max accepts more context: 1M tokens versus 131K.
- Ministral 3B has downloadable open weights; the other is API-only.
Side by side
| Ministral 3B | Qwen3.8 Max | |
|---|---|---|
| Provider | Mistral AI | Alibaba (Qwen) |
| Noometry Index | 26.2 | 56.8 |
| Released | 2024-10-01 | 2026-08-02 |
| Weights | Open | Proprietary |
| Context window | 131K | 1M |
| Max output | 262K | 131K |
| Input $ / M tokens | $0.10 | $2 |
| Output $ / M tokens | $0.10 | $6 |
| Results tracked | 6 | 39 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Not comparable
Ministral 3B: —, Qwen3.8 Max: 53.5 (#29)
| Benchmark | Ministral 3B | Qwen3.8 Max |
|---|---|---|
| DeepSWE | — | 57.5% |
| LMArena WebDev | — | 1674 |
| FrontierSWE | — | 17.8% |
| SciCode | — | 53.2% |
| LMArena Coding | — | 1502 |
Agentic & Tool Use Not comparable
Ministral 3B: —, Qwen3.8 Max: 45.4 (#14)
| Benchmark | Ministral 3B | Qwen3.8 Max |
|---|---|---|
| APEX-Agents | — | 63.3% |
| τ²-bench Banking | — | 55.1% |
| GDP.pdf | — | 23.2% |
Reasoning Qwen3.8 Max leads
Ministral 3B: 18.4 (#282), Qwen3.8 Max: 54.4 (#26)
| Benchmark | Ministral 3B | Qwen3.8 Max |
|---|---|---|
| DTBench | 51.7% | 92% |
| LMCA | 5.5% | 46.2% |
| Epoch Capabilities Index | 118.1 | 156.41 |
| NYT Connections (extended) | — | 88.3% |
| CritPt | — | 20% |
| Chess Puzzles | — | 40% |
| LMArena Hard Prompts | — | 1496 |
| Mystery Game Puzzles | — | 38% |
Math Qwen3.8 Max leads
Ministral 3B: 26.6 (#258), Qwen3.8 Max: 73.2 (#20)
| Benchmark | Ministral 3B | Qwen3.8 Max |
|---|---|---|
| FrontierMath (Tiers 1-3) | — | 74.7% |
| FrontierMath Tier 4 | — | 46.3% |
| OTIS Mock AIME 2024-2025 | — | 100% |
| ProofBench | — | 58% |
| LMArena Math | — | 1499 |
| MATH Level 5 | 14.4% | — |
Knowledge Qwen3.8 Max leads
Ministral 3B: 10.4 (#302), Qwen3.8 Max: 61.7 (#27)
| Benchmark | Ministral 3B | Qwen3.8 Max |
|---|---|---|
| GPQA Diamond | 25.3% | 92.7% |
| SimpleQA Verified | — | 47.3% |
| Vectara Hallucination Rate | 7.3% | — |
| LMArena Expert | — | 1507 |
Multimodal Not comparable
Ministral 3B: —, Qwen3.8 Max: 37.2 (#75)
| Benchmark | Ministral 3B | Qwen3.8 Max |
|---|---|---|
| LMArena Vision | — | 1314 |
| Furniture Assembly | — | 20% |
Multilingual Not comparable
Ministral 3B: —, Qwen3.8 Max: 56.7 (#18)
| Benchmark | Ministral 3B | Qwen3.8 Max |
|---|---|---|
| LMArena Non-English | — | 1472 |
| LMArena Chinese | — | 1538 |
| LMArena French | — | 1503 |
| LMArena German | — | 1483 |
| LMArena Japanese | — | 1467 |
| LMArena Korean | — | 1461 |
| LMArena Russian | — | 1481 |
| LMArena Spanish | — | 1492 |
Instruction Following Not comparable
Ministral 3B: —, Qwen3.8 Max: 77.6 (#17)
| Benchmark | Ministral 3B | Qwen3.8 Max |
|---|---|---|
| LMArena Instruction Following | — | 1479 |
Long Context Not comparable
Ministral 3B: —, Qwen3.8 Max: 45.6 (#31)
| Benchmark | Ministral 3B | Qwen3.8 Max |
|---|---|---|
| LMArena Longer Query | — | 1489 |
Writing & Preference Not comparable
Ministral 3B: —, Qwen3.8 Max: 67.1 (#30)
| Benchmark | Ministral 3B | Qwen3.8 Max |
|---|---|---|
| LMArena Text | — | 1483 |
| LMArena Creative Writing | — | 1479 |
| LMArena Multi-Turn | — | 1489 |
Frequently asked questions
Is Ministral 3B better than Qwen3.8 Max?
Qwen3.8 Max is the stronger model overall, scoring 56.8 to 26.2 on the Noometry Index. Ministral 3B costs 30× less per token, which makes it the better buy when Qwen3.8 Max's lead doesn't matter for your workload.
Which is cheaper, Ministral 3B or Qwen3.8 Max?
Ministral 3B is cheaper. It lists at $0.10 per million input tokens and $0.10 per million output tokens; Qwen3.8 Max lists at $2 and $6.
Which has the bigger context window?
Qwen3.8 Max does, with 1M tokens against 131K.
How many benchmarks do Ministral 3B and Qwen3.8 Max share?
4 benchmarks have published results for both models. Ministral 3B has 6 scored results on Noometry and Qwen3.8 Max has 39.