Currently listed through these providers:
Model details
Codestral (latest)
Codestral is engineered specifically for the demands of modern software development, serving as a specialized tool for code generation, correction, and test creation. Its architecture is built to handle complex programming tasks across more than 80 languages, prioritizing low-latency performance for high-frequency development workflows. By integrating fill-in-the-middle capabilities, the model excels at providing precise code completions and context-aware suggestions, making it a robust companion for developers who require both speed and accuracy in their coding environments.
The model lineage reflects a focus on architectural efficiency and refined tokenization, which significantly accelerates code generation and completion tasks compared to earlier iterations. Designed to support extensive context, it is well-suited for analyzing large codebases and managing complex, multi-file projects. Beyond standard generation, its design supports structured output and tool calling, allowing it to integrate into broader agentic workflows. As a state-of-the-art solution in its class, it provides a balanced approach to reasoning and responsiveness, positioning it as a versatile asset for both automated software development and interactive coding assistance.
Quick Info
Powered by- Provider
- Vercel AI Gateway
- Model key
- mistral/codestral
- Release date
- May 29, 2024
- Last updated
- Jan 4, 2025
- Knowledge cutoff
- 2024-10
- Input modalities
- Output modalities
- Capabilities
Cost
- Input token cost
- $0.30
- Output token cost
- $0.90
Limits
- Output tokens
- 4,096 tokens
- Context window
- 256,000 tokens
Transparent token rates
Compare Codestral (latest) pricing
Rates are shown per one million tokens. Combined means one million input plus one million output tokens.
Latest news about Codestral (latest)
Videos about Codestral (latest)
More models around Codestral (latest)
This exact model name is also listed by 4 other providers.