id: mistralai/devstral-small
canonical_slug: mistralai/devstral-small-2505
hugging_face_id: mistralai/Devstral-Small-2505
name: 'Mistral: Devstral Small'
type: chat
created: 1747837379
description: |-
  Devstral-Small-2505 is a 24B parameter agentic LLM fine-tuned from Mistral-Small-3.1, jointly developed by Mistral AI and All Hands AI for advanced software engineering tasks. It is optimized for codebase exploration, multi-file editing, and integration into coding agents, achieving state-of-the-art results on SWE-Bench Verified (46.8%).

  Devstral supports a 128k context window and uses a custom Tekken tokenizer. It is text-only, with the vision encoder removed, and is suitable for local deployment on high-end consumer hardware (e.g., RTX 4090, 32GB RAM Macs). Devstral is best used in agentic workflows via the OpenHands scaffold and is compatible with inference frameworks like vLLM, Transformers, and Ollama. It is released under the Apache 2.0 license.
context_length: 128000
architecture:
  modality: text->text
  input_modalities:
    - text
  output_modalities:
    - text
  tokenizer: Other
  instruct_type: null
pricing:
  prompt: '0.00000006'
  completion: '0.00000012'
  input_cache_read: ''
  input_cache_write: ''
  request: '0'
  image: '0'
  web_search: '0'
  internal_reasoning: '0'
  unit: 1
  currency: USD
supported_parameters:
  - max_tokens
  - temperature
  - top_p
  - stop
  - frequency_penalty
  - presence_penalty
  - repetition_penalty
  - response_format
  - top_k
  - seed
  - min_p
  - tools
  - tool_choice
  - structured_outputs
model_provider: mistralai