Mistral AI, open weights
Devstral Small 2505
Devstral Small 2505 by Mistral AI ranks 233rd of 354 ranked models on the Noometry Index as of October 2026, with a score of 34.3. Its strongest category is coding, where it ranks 166th. API pricing starts at $0.10 per million input tokens and $0.30 per million output tokens, with a 128K-token context window.
Last verified
Specifications
- Noometry rank
- #233 of 354
- Index score
- 34.3
- Evidence
- Reported 4 results
- Provider
Mistral AI- Released
- May 7, 2025
- Weights
- Open weights
- Reasoning
- No
- Context window
- 128K
- Max output
- 128K
- Input price
- $0.10 / M
- Output price
- $0.30 / M
- Blended price
- $0.15 / M
- Output speed
- 88 tokens/s Kagi
- Value
- #38 of 219
- Knowledge cutoff
- May 2025
- Input
- text
Category scores
Each category score combines every public result we have in that category.
- Coding 38.9
- Reasoning 19.7
Strengths and weaknesses
Categories where Devstral Small 2505 places highest and lowest among the models ranked in each, with its score against that category's median.
Closest competitors
The models ranked just above and below Devstral Small 2505. When scores are this close, price and speed are often the better way to choose.
| Model | Rank | Score | Blended $/M | Speed | |
|---|---|---|---|---|---|
| C4ai Aya Expanse 8b | #229 | 34.9 | — | — | Compare |
| Qwen Max | #230 | 34.7 | $2.80 | — | Compare |
| Claude 3.5 Sonnet | #231 | 34.6 | — | — | Compare |
| Qwen3 Coder Next | #232 | 34.3 | $0.29 | — | Compare |
| Qwen1.5-110B | #234 | 34.2 | — | — | Compare |
| o1-mini | #235 | 34.0 | — | — | Compare |
| Qwen3.5-9B | #236 | 33.8 | $0.11 | — | Compare |
| Codellama 70b Instruct | #237 | 33.7 | — | — | Compare |
Sponsored placements are available on pages like this one. Advertise on Noometry
Benchmark results
Every published result we track, with its source. Bold rows are the ones used for ranking; where several exist we prefer independent runs over self-reported numbers.
Coding
| Benchmark | Score | Position | Setting | Source | Date |
|---|---|---|---|---|---|
| SWE-bench Verified (bash only) | 56.4% | #22 of 39, top 57% | SWE-bench | 2025-12-09 | |
| SciCode | 28.8% | #107 of 121, top 89% | Epoch AI |
Reasoning
| Benchmark | Score | Position | Setting | Source | Date |
|---|---|---|---|---|---|
| Kagi LLM Benchmark | 34.2% | Kagi LLM Benchmark | |||
| Kagi LLM Benchmark | 37.7% | #85 of 99, top 86% | Kagi LLM Benchmark | ||
| CritPt | 0% | #105 of 134, top 79% | Epoch AI |
API pricing by provider
| Route | Input $/M | Output $/M | Cached input $/M | Checked |
|---|---|---|---|---|
| mistral | $0.10 | $0.30 | — | 2026-10-10 |
Compare Devstral Small 2505
- Devstral Small 2505 vs Qwen3 Coder Next
- Devstral Small 2505 vs Qwen1.5-110B
- Devstral Small 2505 vs Claude 3.5 Sonnet
- Devstral Small 2505 vs o1-mini
- Devstral Small 2505 vs Qwen Max
- Devstral Small 2505 vs Qwen3.5-9B
- Devstral Small 2505 vs GPT-6 Astra
- Devstral Small 2505 vs Claude Fable 5.1
- Devstral Small 2505 vs Gemini 3.8 Flash
- Devstral Small 2505 vs Kimi K3
- Devstral Small 2505 vs Grok 4.6
- Devstral Small 2505 vs Qwen3.8 Max
- Devstral Small 2505 vs GLM-5.3
- Devstral Small 2505 vs Muse Spark 1.3
Other Mistral AI models
- Mistral Large 443.1
- Mistral Medium 3.540.2
- Mistral Large 339.1
- Mistral Medium36.3
- Magistral Medium35.2
- Mistral Small33.4
- Pixtral Large32.2
- Mistral Large31.9
Frequently asked questions
How good is Devstral Small 2505?
Devstral Small 2505 by Mistral AI ranks 233rd of 354 ranked models on the Noometry Index as of October 2026, with a score of 34.3. Its strongest category is coding, where it ranks 166th. API pricing starts at $0.10 per million input tokens and $0.30 per million output tokens, with a 128K-token context window.
How much does Devstral Small 2505 cost?
Devstral Small 2505 costs $0.10 per million input tokens and $0.30 per million output tokens on Mistral AI's own API.
What is Devstral Small 2505's context window?
Devstral Small 2505 accepts up to 128K tokens of input and can write up to 128K tokens in one response.
Is Devstral Small 2505 open source?
Yes. Devstral Small 2505's weights are downloadable; check the license for commercial terms.
How fast is Devstral Small 2505?
Devstral Small 2505 generated about 88 output tokens per second in the Kagi LLM Benchmark's timed runs. Speed varies by provider, load and reasoning effort.
What are Devstral Small 2505's strengths and weaknesses?
Relative to other ranked models, Devstral Small 2505 places best in coding and lowest in reasoning.
What is Devstral Small 2505 best at?
Its best category is coding, where it ranks 166th on Noometry.