Model comparison
Devstral Small 2505 vs Qwen3.8 27B
Qwen3.8 27B is the stronger model overall, scoring 46.0 to 34.3 on the Noometry Index. Devstral Small 2505 costs 4.0× less per token, which makes it the better buy when Qwen3.8 27B's lead doesn't matter for your workload.
Last verified . 2 shared benchmarks.
Mistral AI
Alibaba (Qwen)Summary
- They share 2 benchmarks with published results for both. Devstral Small 2505 scores higher in 0 categories and Qwen3.8 27B in 2 categories; 2 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Qwen3.8 27B leads 41.0 to 19.7.
- The biggest single-benchmark swing is SciCode: 28.8% for Devstral Small 2505 and 46.6% for Qwen3.8 27B.
- Devstral Small 2505 is cheaper at $0.10 / $0.30 per million input/output tokens, against $0.04 / $2.30 for Qwen3.8 27B.
- Qwen3.8 27B accepts more context: 262K tokens versus 128K.
Side by side
| Devstral Small 2505 | Qwen3.8 27B | |
|---|---|---|
| Provider | Mistral AI | Alibaba (Qwen) |
| Noometry Index | 34.3 | 46.0 |
| Released | 2025-05-07 | 2026-08-14 |
| Weights | Open | Open |
| Context window | 128K | 262K |
| Max output | 128K | 33K |
| Input $ / M tokens | $0.10 | $0.04 |
| Output $ / M tokens | $0.30 | $2.30 |
| Results tracked | 4 | 31 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen3.8 27B leads
Devstral Small 2505: 38.9 (#166), Qwen3.8 27B: 50.5 (#44)
| Benchmark | Devstral Small 2505 | Qwen3.8 27B |
|---|---|---|
| SciCode | 28.8% | 46.6% |
| SWE-bench Verified (bash only) | 56.4% | — |
| LMArena WebDev | — | 1593 |
| LMArena Coding | — | 1482 |
Agentic & Tool Use Not comparable
Devstral Small 2505: —, Qwen3.8 27B: 32.9 (#57)
| Benchmark | Devstral Small 2505 | Qwen3.8 27B |
|---|---|---|
| APEX-Agents | — | 47.5% |
Reasoning Qwen3.8 27B leads
Devstral Small 2505: 19.7 (#252), Qwen3.8 27B: 41.0 (#54)
| Benchmark | Devstral Small 2505 | Qwen3.8 27B |
|---|---|---|
| CritPt | 0% | 5.4% |
| ARC-AGI-2 | — | 42.4% |
| Kagi LLM Benchmark | 37.7% | — |
| NYT Connections (extended) | — | 54.5% |
| ARC-AGI-1 | — | 87.5% |
| LMArena Hard Prompts | — | 1460 |
| DTBench | — | 88% |
| LMCA | — | 41.4% |
| Surface Evolver Bench | — | 45% |
| Epoch Capabilities Index | — | 149.38 |
Math Not comparable
Devstral Small 2505: —, Qwen3.8 27B: 37.1 (#161)
| Benchmark | Devstral Small 2505 | Qwen3.8 27B |
|---|---|---|
| ProofBench | — | 16% |
| LMArena Math | — | 1456 |
Knowledge Not comparable
Devstral Small 2505: —, Qwen3.8 27B: 41.6 (#109)
| Benchmark | Devstral Small 2505 | Qwen3.8 27B |
|---|---|---|
| LMArena Expert | — | 1482 |
Multimodal Not comparable
Devstral Small 2505: —, Qwen3.8 27B: 41.3 (#37)
| Benchmark | Devstral Small 2505 | Qwen3.8 27B |
|---|---|---|
| LMArena Vision | — | 1271 |
Multilingual Not comparable
Devstral Small 2505: —, Qwen3.8 27B: 53.7 (#60)
| Benchmark | Devstral Small 2505 | Qwen3.8 27B |
|---|---|---|
| LMArena Non-English | — | 1430 |
| LMArena Chinese | — | 1504 |
| LMArena French | — | 1465 |
| LMArena German | — | 1438 |
| LMArena Japanese | — | 1384 |
| LMArena Korean | — | 1393 |
| LMArena Russian | — | 1415 |
| LMArena Spanish | — | 1448 |
Instruction Following Not comparable
Devstral Small 2505: —, Qwen3.8 27B: 75.8 (#53)
| Benchmark | Devstral Small 2505 | Qwen3.8 27B |
|---|---|---|
| LMArena Instruction Following | — | 1439 |
Long Context Not comparable
Devstral Small 2505: —, Qwen3.8 27B: 44.3 (#70)
| Benchmark | Devstral Small 2505 | Qwen3.8 27B |
|---|---|---|
| LMArena Longer Query | — | 1450 |
Writing & Preference Not comparable
Devstral Small 2505: —, Qwen3.8 27B: 65.8 (#43)
| Benchmark | Devstral Small 2505 | Qwen3.8 27B |
|---|---|---|
| LMArena Text | — | 1441 |
| LMArena Creative Writing | — | 1384 |
| EQ-Bench Creative Writing | — | 1671 |
| LMArena Multi-Turn | — | 1441 |
Frequently asked questions
Is Devstral Small 2505 better than Qwen3.8 27B?
Qwen3.8 27B is the stronger model overall, scoring 46.0 to 34.3 on the Noometry Index. Devstral Small 2505 costs 4.0× less per token, which makes it the better buy when Qwen3.8 27B's lead doesn't matter for your workload.
Which is cheaper, Devstral Small 2505 or Qwen3.8 27B?
Devstral Small 2505 is cheaper. It lists at $0.10 per million input tokens and $0.30 per million output tokens; Qwen3.8 27B lists at $0.04 and $2.30.
Is Devstral Small 2505 or Qwen3.8 27B better for coding?
Qwen3.8 27B scores higher on coding benchmarks: 50.5 versus 38.9 in the Noometry coding category.
Which has the bigger context window?
Qwen3.8 27B does, with 262K tokens against 128K.
How many benchmarks do Devstral Small 2505 and Qwen3.8 27B share?
2 benchmarks have published results for both models. Devstral Small 2505 has 4 scored results on Noometry and Qwen3.8 27B has 31.