# Codestral 25.08

> Codestral 25.08 is a proprietary language model from Mistral AI, released on 30 Jul 2025. It accepts text and generates text, with a 256K-token context window and up to 8K output tokens per response. As of 17 Oct 2026, the lowest listed API price is $0.30 per million input tokens and $0.90 per million output tokens (Mistral API), across 3 providers we track.

Source page: https://www.aimodel.directory/models/codestral-25-08
Last verified: 17 Oct 2026

## Specifications

- **Developer:** Mistral AI
- **Release date:** 30 Jul 2025 (source: models.dev, https://github.com/sst/models.dev/blob/dev/providers/mistral/models/codestral-2508.toml)
- **Status:** Generally available (source: models.dev, https://github.com/sst/models.dev/blob/dev/providers/mistral/models/codestral-2508.toml)
- **Weights:** Proprietary (API only) (source: models.dev, https://github.com/sst/models.dev/blob/dev/providers/mistral/models/codestral-2508.toml)
- **Context window:** 256K (256,000 tokens) (source: models.dev, https://github.com/sst/models.dev/blob/dev/providers/mistral/models/codestral-2508.toml)
- **Max output:** 8K tokens (source: models.dev, https://github.com/sst/models.dev/blob/dev/providers/mistral/models/codestral-2508.toml)
- **Knowledge cutoff:** 31 Mar 2025 (source: OpenRouter, https://openrouter.ai/mistralai/codestral-2508)
- **Input:** Text (source: models.dev, https://github.com/sst/models.dev/blob/dev/providers/mistral/models/codestral-2508.toml)
- **Output:** Text (source: models.dev, https://github.com/sst/models.dev/blob/dev/providers/mistral/models/codestral-2508.toml)
- **Reasoning mode:** No (source: models.dev, https://github.com/sst/models.dev/blob/dev/providers/mistral/models/codestral-2508.toml)
- **Tool calling:** Yes (source: models.dev, https://github.com/sst/models.dev/blob/dev/providers/mistral/models/codestral-2508.toml)
- **Structured output:** Yes (source: models.dev, https://github.com/sst/models.dev/blob/dev/providers/mistral/models/codestral-2508.toml)

## API pricing (USD per 1M tokens)

| Provider | Input | Output | Cached input | Context |
|---|---|---|---|---|
| Mistral API | $0.30 | $0.90 | $0.03 | 256K |
| Google Vertex AI | $0.30 | $0.90 | – | 128K |
| OpenRouter | $0.30 | $0.90 | $0.03 | 256K |

## FAQ

### What is the context window of Codestral 25.08?

Codestral 25.08 has a context window of 256,000 tokens (256K) and can generate up to 8,192 tokens in a single response, according to models.dev.

### How much does the Codestral 25.08 API cost?

Through Mistral API, Codestral 25.08 costs $0.30 per million input tokens and $0.90 per million output tokens. Prices last verified 17 Oct 2026.

### Is Codestral 25.08 open source?

No. Codestral 25.08 is a proprietary model; Mistral AI has not released its weights. It is available through APIs including Mistral API, Google Vertex AI, OpenRouter.

### When was Codestral 25.08 released?

Mistral AI released Codestral 25.08 on 30 Jul 2025, according to models.dev.

### Which providers offer Codestral 25.08?

We track Codestral 25.08 on 3 providers: Mistral API, Google Vertex AI, OpenRouter.

---
Data licensed CC BY 4.0. Cite as: AI Model Directory, "Codestral 25.08", https://www.aimodel.directory/models/codestral-25-08