# Mistral: Codestral 2508 (batch)

Provider: mistralai  
Category: llm  
Model ID: `mistralai/codestral-2508:batch`

Mistral's cutting-edge language model for coding released end of July 2025. Codestral specializes in low-latency, high-frequency tasks such as fill-in-the-middle (FIM), code correction and test generation.

[Blog Post](https://mistral.ai/news/codestral-25-08)

## Specs

- Context length: 256000 tokens
- Max output: 204800 tokens
- Modalities: text, file
- Released: 2025-08-01

## Pricing

- Input: $0.15 per million tokens
- Output: $0.45 per million tokens

## Providers

- **mistralai** — ctx 256000, input $0.15/M, output $0.45/M

## Verdict

Codestral 2508 batch is Mistral's specialized coding model optimized for asynchronous workloads where latency doesn't matter. The 256K context window handles entire codebases, and at $0.15/$0.45 per Mtok it undercuts real-time alternatives by roughly 50%. The batch-only constraint means you're queuing requests with delayed responses—fine for overnight refactoring jobs or CI/CD analysis, but unusable for interactive coding sessions. Reach for this when you're processing large codebases in bulk and can wait minutes to hours for results.

---
Last verified: 2026-10-19T02:30:34.239Z  
Canonical URL: https://switchy.build/directory/models/codestral-2508-batch