Model comparison
Ministral 8B vs Qwen3.5 397B-A17B
Qwen3.5 397B-A17B is the stronger model overall, scoring 46.0 to 28.2 on the Noometry Index. Ministral 8B costs 9.0× less per token, which makes it the better buy when Qwen3.5 397B-A17B's lead doesn't matter for your workload.
Last verified . 14 shared benchmarks.
Mistral AI
Alibaba (Qwen)Summary
- They share 14 benchmarks with published results for both. Ministral 8B scores higher in 0 categories and Qwen3.5 397B-A17B in 9 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Qwen3.5 397B-A17B leads 53.3 to 12.6.
- The biggest single-benchmark swing is GPQA Diamond: 27.1% for Ministral 8B and 86.4% for Qwen3.5 397B-A17B.
- Ministral 8B is cheaper at $0.15 / $0.15 per million input/output tokens, against $0.60 / $3.60 for Qwen3.5 397B-A17B.
Side by side
| Ministral 8B | Qwen3.5 397B-A17B | |
|---|---|---|
| Provider | Mistral AI | Alibaba (Qwen) |
| Noometry Index | 28.2 | 46.0 |
| Released | 2024-10-01 | 2026-02-01 |
| Weights | Open | Open |
| Context window | 262K | 262K |
| Max output | 262K | 66K |
| Input $ / M tokens | $0.15 | $0.60 |
| Output $ / M tokens | $0.15 | $3.60 |
| Results tracked | 17 | 36 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen3.5 397B-A17B leads
Ministral 8B: 35.0 (#230), Qwen3.5 397B-A17B: 42.0 (#114)
| Benchmark | Ministral 8B | Qwen3.5 397B-A17B |
|---|---|---|
| LMArena Coding | 1202 | 1465 |
| LMArena WebDev | — | 1400 |
Agentic & Tool Use Qwen3.5 397B-A17B leads
Ministral 8B: 16.4 (#148), Qwen3.5 397B-A17B: 33.3 (#53)
| Benchmark | Ministral 8B | Qwen3.5 397B-A17B |
|---|---|---|
| APEX-Agents | — | 24.9% |
| Berkeley Function Calling Leaderboard | 11.1% | — |
| τ²-bench Airline | — | 81.5% |
| τ²-bench Banking | — | 9.8% |
| τ²-bench Retail | — | 84.4% |
| τ²-bench Telecom | — | 97.8% |
Reasoning Qwen3.5 397B-A17B leads
Ministral 8B: 18.4 (#281), Qwen3.5 397B-A17B: 34.5 (#70)
| Benchmark | Ministral 8B | Qwen3.5 397B-A17B |
|---|---|---|
| LMArena Hard Prompts | 1191 | 1448 |
| DTBench | 45.7% | 87.5% |
| Kagi LLM Benchmark | — | 73.7% |
| NYT Connections (extended) | — | 58.9% |
| Chess Puzzles | — | 13% |
| Thematic Generalization | — | 65.1% |
| Mystery Game Puzzles | — | 18% |
| LMCA | — | 37.9% |
| Epoch Capabilities Index | — | 146.65 |
Math Qwen3.5 397B-A17B leads
Ministral 8B: 25.7 (#267), Qwen3.5 397B-A17B: 46.1 (#73)
| Benchmark | Ministral 8B | Qwen3.5 397B-A17B |
|---|---|---|
| LMArena Math | 1188 | 1454 |
| FrontierMath (Tiers 1-3) | — | 31.2% |
| OTIS Mock AIME 2024-2025 | — | 88.9% |
| MATH Level 5 | 14.9% | — |
Knowledge Qwen3.5 397B-A17B leads
Ministral 8B: 12.6 (#297), Qwen3.5 397B-A17B: 53.3 (#58)
| Benchmark | Ministral 8B | Qwen3.5 397B-A17B |
|---|---|---|
| GPQA Diamond | 27.1% | 86.4% |
| LMArena Expert | 1170 | 1462 |
| Vectara Hallucination Rate | 7.4% | — |
Multimodal Not comparable
Ministral 8B: —, Qwen3.5 397B-A17B: 40.7 (#44)
| Benchmark | Ministral 8B | Qwen3.5 397B-A17B |
|---|---|---|
| LMArena Vision | — | 1263 |
Multilingual Qwen3.5 397B-A17B leads
Ministral 8B: 35.1 (#247), Qwen3.5 397B-A17B: 53.7 (#59)
| Benchmark | Ministral 8B | Qwen3.5 397B-A17B |
|---|---|---|
| LMArena Non-English | 1165 | 1430 |
| LMArena Chinese | 1193 | 1500 |
| LMArena Russian | 1195 | 1429 |
| LMArena French | — | 1461 |
| LMArena German | — | 1447 |
| LMArena Japanese | — | 1426 |
| LMArena Korean | — | 1384 |
| LMArena Spanish | — | 1441 |
Instruction Following Qwen3.5 397B-A17B leads
Ministral 8B: 60.5 (#250), Qwen3.5 397B-A17B: 75.0 (#77)
| Benchmark | Ministral 8B | Qwen3.5 397B-A17B |
|---|---|---|
| LMArena Instruction Following | 1161 | 1424 |
Long Context Qwen3.5 397B-A17B leads
Ministral 8B: 36.7 (#227), Qwen3.5 397B-A17B: 44.1 (#74)
| Benchmark | Ministral 8B | Qwen3.5 397B-A17B |
|---|---|---|
| LMArena Longer Query | 1212 | 1442 |
Writing & Preference Qwen3.5 397B-A17B leads
Ministral 8B: 39.6 (#246), Qwen3.5 397B-A17B: 62.3 (#79)
| Benchmark | Ministral 8B | Qwen3.5 397B-A17B |
|---|---|---|
| LMArena Text | 1191 | 1438 |
| LMArena Creative Writing | 1175 | 1401 |
| LMArena Multi-Turn | 1166 | 1446 |
| EQ-Bench Creative Writing | — | 1478 |
Frequently asked questions
Is Ministral 8B better than Qwen3.5 397B-A17B?
Qwen3.5 397B-A17B is the stronger model overall, scoring 46.0 to 28.2 on the Noometry Index. Ministral 8B costs 9.0× less per token, which makes it the better buy when Qwen3.5 397B-A17B's lead doesn't matter for your workload.
Which is cheaper, Ministral 8B or Qwen3.5 397B-A17B?
Ministral 8B is cheaper. It lists at $0.15 per million input tokens and $0.15 per million output tokens; Qwen3.5 397B-A17B lists at $0.60 and $3.60.
Is Ministral 8B or Qwen3.5 397B-A17B better for coding?
Qwen3.5 397B-A17B scores higher on coding benchmarks: 42.0 versus 35.0 in the Noometry coding category.
Which has the bigger context window?
Both accept 262K tokens.
How many benchmarks do Ministral 8B and Qwen3.5 397B-A17B share?
14 benchmarks have published results for both models. Ministral 8B has 17 scored results on Noometry and Qwen3.5 397B-A17B has 36.