Codestral 2508
Mistral· released 1 Aug 2025
mistralai/codestral-2508- Input / 1M tokens
- $0.3
- Output / 1M tokens
- $0.9
Longest answer: 205K tokens
About
Codestral 2508 is a Mistral language model focused on coding tasks. Released on 2025-08-01, it is described as suited to low-latency, high-frequency work such as fill-in-the-middle completion, code correction, and test generation. It accepts text and files, returns text, and supports tools. Its context window is 256,000 tokens, and its maximum output is 204,800 tokens. The model ID is mistralai/codestral-2508. Listed pricing is USD 0.3 per 1M input tokens and USD 0.9 per 1M output tokens. For teams choosing a coding model, these details connect its stated use cases with practical constraints: input formats, tool support, context capacity, maximum output, and per-token costs. The listing does not specify reasoning support, so builders should assess the capabilities that are explicitly stated here when deciding whether it fits a coding workflow.
Who it is for
Codestral 2508 suits teams handling frequent coding tasks such as fill-in-the-middle completion, code correction, or test generation. File input and tool support may fit workflows that use those capabilities.
What is good
- Focused on fill-in-the-middle coding tasks.
- Supports code correction and test generation.
- Accepts text and file inputs.
- Maximum output is 204,800 tokens.
What to know first
- Input costs USD 0.3 per 1M tokens.
- Output costs USD 0.9 per 1M tokens.
Verdict
Codestral 2508 is oriented toward frequent coding tasks, with file input, tool support, and a 256,000-token context window. Check that its stated task focus and token costs suit your coding workflow.
Details
- Lab
- Mistralopenrouter.ai · 3 Oct 2026
- Context
- 256Kopenrouter.ai · 3 Oct 2026
- Input price
- $0.3 / 1Mopenrouter.ai · 3 Oct 2026
- Output price
- $0.9 / 1Mopenrouter.ai · 3 Oct 2026
- Max output
- 204,800 tokensopenrouter.ai · 3 Oct 2026
- Inputs
- text, fileopenrouter.ai · 3 Oct 2026
- Open weights
- Noopenrouter.ai · 3 Oct 2026
More longest-context models
See the list
Listed on Inferse
- Longest-Context Models in 2026466 listed
- Newest Models in 2026466 listed
- Cheapest Language Models in 2026437 listed
- Models with Tool Calling in 2026395 listed
- Mistral Models: Pricing per 1M Tokens and Context (2026)23 listed
Sources
- openrouter.ai/mistralai/codestral-2508· checked 3 Oct 2026