Devstral 2 2512
Devstral 2 2512 is an open-source model developed by Mistral AI specifically for agentic coding tasks. It features a 256K context window and is designed to handle complex programming workflows and file-based inputs.
Devstral 2 2512 specs and pricing
- Provider: Mistral
- Released: December 9, 2025
- Context window: 262K tokens
- Max output: 210K tokens
- Input: text, files
- Output: text
- Price: $0.40 input / $2.00 output per 1M tokens
- Tool calling: yes
What Devstral 2 2512 is good at
- Automating complex software development tasks
- Analyzing large codebases and project documentation
- Executing multi-step programming workflows via tool calling
To-dos to try with Devstral 2 2512
- Analyze this entire repository and identify potential security vulnerabilities in the authentication module.
- Read these five technical specification files and generate a corresponding class structure in Python.
- Review the provided error logs and suggest a refactoring plan for the identified memory leak.
FAQ
- What is Devstral 2 2512 primarily designed to do? It is a specialized model built by Mistral AI to function as an agent for coding and software development tasks.
- Can this model process files? Yes, the model is designed to accept both text and file inputs to assist with development workflows.
- How much does Devstral 2 2512 cost? Devstral 2 2512 costs $0.40 per million input tokens and $2.00 per million output tokens through OpenRouter. In todo.is a typical agent step costs about 1.4 credits ($1 = 100 credits).
- What is the context window of Devstral 2 2512? Devstral 2 2512 can read up to 262K tokens at once and write up to 210K tokens in one reply.
- Can I use Devstral 2 2512 without coding? Yes. In todo.is you pick Devstral 2 2512 as your agent's model and write to-dos in plain words, on the web, the Android app, WhatsApp or Telegram. No API key needed: credits are included, or you can bring your own key.
JavaScript is required to use the todo.is app.