Model Information

Display Name: Codestral

API Model ID: mistralai/codestral

Category: Text To Text

Description: Codestral is Mistral AI's specialized code generation model, optimized for code completion, generation, and software engineering tasks with a massive 256K context window. **Key Features:** - 256K token context window - Specialized for code tasks - Fill-in-the-middle (FIM) support - 80+ programming languages - Low latency inference - IDE integration ready **Capabilities:** - Code completion and generation - Code explanation and documentation - Bug fixing and debugging - Code refactoring - Test generation - Multi-file context understanding **Best For:** - IDE code completion - Developer tools - Code review automation - Documentation generation - Large codebase analysis **Technical Specs:** - Parameters: 22B - Context: 256K tokens - Supports: Python, JavaScript, TypeScript, Java, C++, Go, Rust, and 80+ languages - License: Commercial

Context Window: 262,144 tokens

Max Output: 32,768 tokens

How to Use This Model

To use Codestral via the HInow.ai API, use the model ID: mistralai/codestral

API Request Example (Chat/Text)

POST https://api.hinow.ai/v1/chat/completions
Authorization: Bearer YOUR_API_KEY
Content-Type: application/json

{
  "model": "mistralai/codestral",
  "messages": [
    {"role": "user", "content": "Your message here"}
  ]
}
              

Pricing

• input: $0.27
• output: $0.81

Available Parameters

• temperature: Controls randomness (lower for code) (Options: 0, 0.2, 0.4, 0.6, 0.8)
• top_p: Nucleus sampling (Options: 0.1, 0.5, 0.7, 0.9, 0.95)
• max_tokens: Max tokens (Options: 256, 512, 1024, 2048, 4096, 8192)
• stop: Stop sequences

Quick Reference

To use this model, set: "model": "mistralai/codestral"

Featured: Yes

Documentation: https://hinow.ai/models/mistralai/codestral

API Endpoint: https://api.hinow.ai/v1