Model Library
MistralReleased December 9, 2025Devstral Small 2
Mistral · Released December 9, 2025
Intelligence score
17
At a glance
- Intelligence
- 17
- Speed
- 67.8tokens/sec
- Latency
- 1.20stime to first token
- Input price
- Not published
- Output price
- Not published
- Context window
- 260Ktokens
- Parameters
- 24billions
Summary
Comparison summary
Devstral Small 2 scores 17 on the Artificial Analysis Intelligence Index, placing it well above average among other open weight non-reasoning models of similar size (median: 6). Devstral Small 2 generates output at 67.8 tokens per second (based on Mistral's API), which is at the lower end compared to other open weight non-reasoning models of similar size (median: 102.7 t/s).
Specifications
Technical specifications
| Model name | Devstral Small 2 |
|---|---|
| Provider | Mistral |
| Release date | December 9, 2025 |
| Intelligence score | 17 |
| Speed | 67.8 tokens/sec |
| Latency (TTFT) | 1.20s |
| Input price | Not published |
| Output price | Not published |
| Context window | 260,000 tokens |
| Parameters | 24B |
| Reasoning | No |
| Input modalities | text, image |
| Output modalities | — |
| Open weights | Yes |
| License | Apache 2 |
| API availability | — |
| Source | Model Intelligence legacy profile: devstral-small-2 |
FAQ · 18
Frequently asked questions
Devstral Small 2 was released on December 9, 2025.
Devstral Small 2 was created by Mistral.
Devstral Small 2 scores 17 on the Artificial Analysis Intelligence Index, placing it well above average among other open weight non-reasoning models of similar size (median: 6).
Devstral Small 2 generates output at 67.8 tokens per second (based on Mistral's API), which is at the lower end compared to other open weight non-reasoning models of similar size (median: 102.7 t/s).
Devstral Small 2 has a time to first token (TTFT) of 1.20s (based on Mistral's API), which is better than average compared to other open weight non-reasoning models of similar size (median: 1.54s).
When evaluated on the Intelligence Index, Devstral Small 2 generated 9.5M output tokens, which is better than average compared to other open weight non-reasoning models of similar size (median: 9.5M).
No, Devstral Small 2 is not a reasoning model. It provides direct responses without extended chain-of-thought reasoning.
Devstral Small 2 supports text and image input.
Devstral Small 2 supports text output.
Yes, Devstral Small 2 supports image input and can analyze, describe, and answer questions about images.
Yes, Devstral Small 2 is multimodal. It can process text and image input and generate text output.
Devstral Small 2 has a context window of 260k tokens. This determines how much text and conversation history the model can process in a single request.
Yes, Devstral Small 2 is open weights. The model weights are publicly available and can be downloaded for self-hosting.
Devstral Small 2 has 24 billion parameters.
Devstral Small 2 is released under the Apache 2.0 license. This license allows commercial use.
Devstral Small 2 achieves a score of 17 on the Artificial Analysis Intelligence Index. This composite benchmark evaluates models across reasoning, knowledge, mathematics, and coding.
Yes, Devstral Small 2 is available via API through 1 provider.
Devstral Small 2 is available through 1 API provider.