Activefastllm Proprietary

Codestral 2025

by Mistral AI· Released May 2025· Cutoff May 2025

Codestral 2025 is Mistral AI's latest code-focused model, optimized for code generation, completion, and reasoning. It offers a 256K context window and is designed for high-performance coding tasks with low latency. This model is part of Mistral's specialized lineup for developers.

Official Site API Docs

Input cost

$0.20 per 1M tokens

Output cost

$0.60 per 1M tokens

Context window

256K tokens

Max output

—

Modalities

text

License

proprietary

Capabilities

Code GenerationCode CompletionCode ReasoningFunction CallingStreamingJSON Mode

Best For

High-performance code generation and completion with a large context window.

Strengths

Excellent code generation quality
Very large 256K context window
Low latency and fast inference
Competitive pricing

Limitations

Primarily focused on code, less suited for general text tasks
No vision or multimodal capabilities
May not match frontier models on complex reasoning benchmarks

Use Cases

Automated code generation

Code completion in IDEs

Code review and refactoring

Documentation generation from code

Bug fixing and debugging assistance

Learning and tutoring programming

Generating unit tests

Improvements Over Previous Model

Context window increased from 32K to 256K tokens
Improved code generation accuracy and reduced hallucinations
Lower pricing compared to previous Codestral versions
Faster inference speed
Enhanced function calling capabilities

Back to all models