codestral-latest

codestral-latest is a efficient-tier language model from Mistral with 22B parameters. codestral-latest is a proprietary model available exclusively through Mistral's API. The Codestral family is Mistral's model series for this capability tier.

codestral-latest is priced at $0.3/1M input tokens and $0.9/1M output tokens. Output tokens are 3× more expensive than input tokens — this ratio reflects the sequential nature of autoregressive generation, which requires a separate forward pass per output token versus parallel processing of all input tokens simultaneously. At a typical usage pattern of 1,000 input tokens and 500 output tokens per call, codestral-latest costs approximately $0.75 per 1,000 API calls. Mistral does not publish a prompt caching rate for this model, so repeated long contexts are billed at the standard $0.3/1M input rate.

Its 262K-token context window is in the top tier for production models, enabling full-document analysis, long multi-turn conversations, and large codebase understanding without chunking. Median latency (P50 time-to-first-token) is approximately 350ms — suitable for interactive applications where sub-second response is acceptable. Throughput is approximately 200 tokens/second — fast enough to stream responses at reading speed with headroom for concurrent requests. Maximum output length is 8K tokens — adequate for most generation tasks including essays, code functions, and detailed responses.

codestral-latest is best suited to coding, cost. Supported capabilities include function calling / tool use for agentic workflows, streaming token output for real-time UIs. Function calling support makes it suitable for building AI agents, structured data extraction pipelines, and any workflow requiring the model to invoke external APIs or return typed JSON. When evaluating this model for a specific workload, the most important factors are: context window (does your input fit?), output price (how long are your generations?), and latency (is this interactive or batch?). The pricing table above shows how codestral-latest compares to similar models across these dimensions.

codestral-latest sits in the efficient tier — optimized for speed and cost while maintaining solid quality, ideal for high-volume or latency-sensitive workloads. Unlike open-weights alternatives, codestral-latest is only available via Mistral's API — there is no self-hosted option, which simplifies deployment but creates vendor dependency. When comparing codestral-latest to alternatives, use the similar models section on this page to find models at comparable price points with different capability tradeoffs. The intelligence score (62/100) reflects aggregate benchmark performance — a useful proxy for general capability, though real-world performance varies significantly by task type. Always test with your specific prompts and use case before committing to a model for production.