Model comparison
Devstral Small 2505 vs Qwen3 14B
Qwen3 14B is the stronger model overall, scoring 35.5 to 34.3 on the Noometry Index. Devstral Small 2505 costs 4.1× less per token, which makes it the better buy when Qwen3 14B's lead doesn't matter for your workload.
Last verified . 3 shared benchmarks.
Mistral AI
Alibaba (Qwen)Summary
- They share 3 benchmarks with published results for both. Devstral Small 2505 scores higher in 2 categories and Qwen3 14B in 0 categories; 2 gaps are clear of the uncertainty.
- The biggest single-benchmark swing is Kagi LLM Benchmark: 37.7% for Devstral Small 2505 and 49.1% for Qwen3 14B.
- Devstral Small 2505 is cheaper at $0.10 / $0.30 per million input/output tokens, against $0.35 / $1.40 for Qwen3 14B.
- Qwen3 14B accepts more context: 131K tokens versus 128K.
Side by side
| Devstral Small 2505 | Qwen3 14B | |
|---|---|---|
| Provider | Mistral AI | Alibaba (Qwen) |
| Noometry Index | 34.3 | 35.5 |
| Released | 2025-05-07 | 2025-04 |
| Weights | Open | Open |
| Context window | 128K | 131K |
| Max output | 128K | 8K |
| Input $ / M tokens | $0.10 | $0.35 |
| Output $ / M tokens | $0.30 | $1.40 |
| Results tracked | 4 | 12 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Devstral Small 2505 leads
Devstral Small 2505: 38.9 (#166), Qwen3 14B: 37.3 (#195)
| Benchmark | Devstral Small 2505 | Qwen3 14B |
|---|---|---|
| SciCode | 28.8% | 31.6% |
| SWE-bench Verified (bash only) | 56.4% | — |
Agentic & Tool Use Not comparable
Devstral Small 2505: —, Qwen3 14B: 29.6 (#83)
| Benchmark | Devstral Small 2505 | Qwen3 14B |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 41% |
Reasoning Devstral Small 2505 leads
Devstral Small 2505: 19.7 (#252), Qwen3 14B: 18.5 (#280)
| Benchmark | Devstral Small 2505 | Qwen3 14B |
|---|---|---|
| Kagi LLM Benchmark | 37.7% | 49.1% |
| CritPt | 0% | 0% |
| Chess Puzzles | — | 4% |
| DTBench | — | 64% |
| LMCA | — | 18.2% |
| Epoch Capabilities Index | — | 138.23 |
Math Not comparable
Devstral Small 2505: —, Qwen3 14B: 38.6 (#133)
| Benchmark | Devstral Small 2505 | Qwen3 14B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | — | 66.4% |
Knowledge Not comparable
Devstral Small 2505: —, Qwen3 14B: 39.3 (#134)
| Benchmark | Devstral Small 2505 | Qwen3 14B |
|---|---|---|
| GPQA Diamond | — | 63.8% |
| Vectara Hallucination Rate | — | 5.4% |
Long Context Not comparable
Devstral Small 2505: —, Qwen3 14B: 38.1 (#204)
| Benchmark | Devstral Small 2505 | Qwen3 14B |
|---|---|---|
| Fiction.LiveBench | — | 62.5% |
Frequently asked questions
Is Devstral Small 2505 better than Qwen3 14B?
Qwen3 14B is the stronger model overall, scoring 35.5 to 34.3 on the Noometry Index. Devstral Small 2505 costs 4.1× less per token, which makes it the better buy when Qwen3 14B's lead doesn't matter for your workload.
Which is cheaper, Devstral Small 2505 or Qwen3 14B?
Devstral Small 2505 is cheaper. It lists at $0.10 per million input tokens and $0.30 per million output tokens; Qwen3 14B lists at $0.35 and $1.40.
Is Devstral Small 2505 or Qwen3 14B better for coding?
Devstral Small 2505 scores higher on coding benchmarks: 38.9 versus 37.3 in the Noometry coding category.
Which has the bigger context window?
Qwen3 14B does, with 131K tokens against 128K.
How many benchmarks do Devstral Small 2505 and Qwen3 14B share?
3 benchmarks have published results for both models. Devstral Small 2505 has 4 scored results on Noometry and Qwen3 14B has 12.