Model comparison
Devstral Small 2505 vs Qwen3.5 397B-A17B
Qwen3.5 397B-A17B is the stronger model overall, scoring 45.7 to 33.9 on the Noometry Index. Devstral Small 2505 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 . 1 shared benchmarks.
Mistral AI
Alibaba (Qwen)Summary
- They share 1 benchmark with published results for both. Devstral Small 2505 scores higher in 0 categories and Qwen3.5 397B-A17B in 2 categories; 2 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Qwen3.5 397B-A17B leads 33.6 to 19.1.
- The biggest single-benchmark swing is Kagi LLM Benchmark: 37.7% for Devstral Small 2505 and 73.7% for Qwen3.5 397B-A17B.
- Devstral Small 2505 is cheaper at $0.10 / $0.30 per million input/output tokens, against $0.60 / $3.60 for Qwen3.5 397B-A17B.
- Qwen3.5 397B-A17B accepts more context: 262K tokens versus 128K.
Side by side
| Devstral Small 2505 | Qwen3.5 397B-A17B | |
|---|---|---|
| Provider | Mistral AI | Alibaba (Qwen) |
| Noometry Index | 33.9 | 45.7 |
| Released | 2025-05-07 | 2026-02-01 |
| Weights | Open | Open |
| Context window | 128K | 262K |
| Max output | 128K | 66K |
| Input $ / M tokens | $0.10 | $0.60 |
| Output $ / M tokens | $0.30 | $3.60 |
| Results tracked | 4 | 36 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen3.5 397B-A17B leads
Devstral Small 2505: 38.6 (#160), Qwen3.5 397B-A17B: 41.7 (#108)
| Benchmark | Devstral Small 2505 | Qwen3.5 397B-A17B |
|---|---|---|
| SWE-bench Verified (bash only) | 56.4% | — |
| LMArena WebDev | — | 1400 |
| SciCode | 28.8% | — |
| LMArena Coding | — | 1464 |
Agentic & Tool Use Not comparable
Devstral Small 2505: —, Qwen3.5 397B-A17B: 33.2 (#51)
| Benchmark | Devstral Small 2505 | Qwen3.5 397B-A17B |
|---|---|---|
| APEX-Agents | — | 24.9% |
| 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
Devstral Small 2505: 19.1 (#243), Qwen3.5 397B-A17B: 33.6 (#62)
| Benchmark | Devstral Small 2505 | Qwen3.5 397B-A17B |
|---|---|---|
| Kagi LLM Benchmark | 37.7% | 73.7% |
| NYT Connections (extended) | — | 58.9% |
| CritPt | 0% | — |
| Chess Puzzles | — | 13% |
| Thematic Generalization | — | 65.1% |
| LMArena Hard Prompts | — | 1448 |
| Mystery Game Puzzles | — | 18% |
| DTBench | — | 87.5% |
| LMCA | — | 37.9% |
| Epoch Capabilities Index | — | 146.64 |
Math Not comparable
Devstral Small 2505: —, Qwen3.5 397B-A17B: 44.9 (#64)
| Benchmark | Devstral Small 2505 | Qwen3.5 397B-A17B |
|---|---|---|
| FrontierMath (Tiers 1-3) | — | 31.2% |
| OTIS Mock AIME 2024-2025 | — | 88.9% |
| LMArena Math | — | 1455 |
Knowledge Not comparable
Devstral Small 2505: —, Qwen3.5 397B-A17B: 53.3 (#53)
| Benchmark | Devstral Small 2505 | Qwen3.5 397B-A17B |
|---|---|---|
| GPQA Diamond | — | 86.4% |
| LMArena Expert | — | 1461 |
Multimodal Not comparable
Devstral Small 2505: —, Qwen3.5 397B-A17B: 40.0 (#42)
| Benchmark | Devstral Small 2505 | Qwen3.5 397B-A17B |
|---|---|---|
| LMArena Vision | — | 1262 |
Multilingual Not comparable
Devstral Small 2505: —, Qwen3.5 397B-A17B: 54.0 (#56)
| Benchmark | Devstral Small 2505 | Qwen3.5 397B-A17B |
|---|---|---|
| LMArena Non-English | — | 1431 |
| LMArena Chinese | — | 1501 |
| LMArena French | — | 1461 |
| LMArena German | — | 1447 |
| LMArena Japanese | — | 1427 |
| LMArena Korean | — | 1382 |
| LMArena Russian | — | 1429 |
| LMArena Spanish | — | 1443 |
Instruction Following Not comparable
Devstral Small 2505: —, Qwen3.5 397B-A17B: 75.2 (#73)
| Benchmark | Devstral Small 2505 | Qwen3.5 397B-A17B |
|---|---|---|
| LMArena Instruction Following | — | 1424 |
Long Context Not comparable
Devstral Small 2505: —, Qwen3.5 397B-A17B: 44.1 (#71)
| Benchmark | Devstral Small 2505 | Qwen3.5 397B-A17B |
|---|---|---|
| LMArena Longer Query | — | 1442 |
Writing & Preference Not comparable
Devstral Small 2505: —, Qwen3.5 397B-A17B: 62.5 (#74)
| Benchmark | Devstral Small 2505 | Qwen3.5 397B-A17B |
|---|---|---|
| LMArena Text | — | 1438 |
| LMArena Creative Writing | — | 1401 |
| EQ-Bench Creative Writing | — | 1478 |
| LMArena Multi-Turn | — | 1447 |
Frequently asked questions
Is Devstral Small 2505 better than Qwen3.5 397B-A17B?
Qwen3.5 397B-A17B is the stronger model overall, scoring 45.7 to 33.9 on the Noometry Index. Devstral Small 2505 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, Devstral Small 2505 or Qwen3.5 397B-A17B?
Devstral Small 2505 is cheaper. It lists at $0.10 per million input tokens and $0.30 per million output tokens; Qwen3.5 397B-A17B lists at $0.60 and $3.60.
Is Devstral Small 2505 or Qwen3.5 397B-A17B better for coding?
Qwen3.5 397B-A17B scores higher on coding benchmarks: 41.7 versus 38.6 in the Noometry coding category.
Which has the bigger context window?
Qwen3.5 397B-A17B does, with 262K tokens against 128K.
How many benchmarks do Devstral Small 2505 and Qwen3.5 397B-A17B share?
1 benchmark has published results for both models. Devstral Small 2505 has 4 scored results on Noometry and Qwen3.5 397B-A17B has 36.