Mistral: Devstral Small 1.1

Mistral AIID: mistralai/devstral-small

Devstral Small 1.1 is a 24B parameter open-weight language model for software engineering agents, developed by Mistral AI in collaboration with All Hands AI. Finetuned from Mistral Small 3.1 and released under the Apache 2.0 license, it features a 128k token context window and supports both Mistral-style function calling and XML output formats. Designed for agentic coding workflows, Devstral Small 1.1 is optimized for tasks such as codebase exploration, multi-file edits, and integration into autonomous development agents like OpenHands and Cline. It achieves 53.6% on SWE-Bench Verified, surpassing all other open models on this benchmark, while remaining lightweight enough to run on a single 4090 GPU or Apple silicon machine. The model uses a Tekken tokenizer with a 131k vocabulary and is deployable via vLLM, Transformers, Ollama, LM Studio, and other OpenAI-compatible runtimes.

Pricing per 1M Tokens

Input (Prompt)$0.10
Output (Completion)$0.30
Cache Read$0.01
Cache WriteFree
ImageN/A

Specifications

Context Length131K
Max Output TokensN/A
Input ModalitiesText
Output ModalitiesText
TokenizerMistral
Instruct TypeN/A
Top Provider Context131K
Top Provider Max OutputN/A
ModeratedNo

Compare this model

See how Mistral: Devstral Small 1.1 stacks up against other models.

vs Mistral: Mistral Small 4vs Mistral: Ministral 3 8B 2512vs Mistral: Ministral 3 3B 2512vs Mistral: Mistral Large 3 2512vs Mistral: Mistral Small Creative

More from Mistral AI

Mistral: Mistral Small 4
Input$0.15
Context262K
Mistral: Ministral 3 8B 2512
Input$0.15
Context262K
Mistral: Ministral 3 3B 2512
Input$0.10
Context131K
Mistral: Mistral Large 3 2512
Input$0.50
Context262K
Mistral: Mistral Small Creative
Input$0.10
Context33K

Last updated: March 23, 2026

First tracked: March 23, 2026