Home › Providers › mistralai › Mistral: Codestral 2508
Mistral: Codestral 2508 API Pricing
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)
How much does Mistral: Codestral 2508 cost for a real workload?
At 10M input + 2M output tokens per month, Mistral: Codestral 2508 costs about $4.8 ($3 input + $1.8 output). Try your own volumes in the cost calculator.
What is the official Mistral: Codestral 2508 price?
| Official (mistralai) | Catalog (OpenRouter) | |
|---|---|---|
| Input /1M | $0.3 | $0.3 |
| Output /1M | $0.9 | $0.9 |
| Cached input /1M | $0.03 | $0.03 |
✓ Matches the official list price. Source: mistralai official pricing page, checked 2026-10-21.
What are cheaper alternatives to Mistral: Codestral 2508?
- Qwen: Qwen3 Next 80B A3B Instruct — $0.09 in / $1.1 out per 1M
- Qwen: Qwen3 Coder Flash — $0.195 in / $0.975 out per 1M
- DeepSeek: DeepSeek V4 Pro 0423 — $0.3835 in / $0.767 out per 1M
- Qwen: Qwen3.7 Plus (batch) — $0.23 in / $0.92 out per 1M
- Mancer: Weaver (alpha) — $0.4 in / $0.75 out per 1M
- Qwen: Qwen3.5-35B-A3B — $0.14 in / $1 out per 1M
Where to run Mistral: Codestral 2508
AI/ML API: one API key for 400+ models, including most models tracked here · Novita AI: pay-as-you-go APIs for open models (DeepSeek, Qwen, Llama and more) plus GPU cloud. Prices there follow each platform's own catalog and can differ from the list prices tracked here.
Run it on AI/ML API →Run it on Novita AI →
Disclosure: links to these platforms may be affiliate links — if you sign up or buy through them, LLMPrices may earn a commission at no extra cost to you.
Compare Mistral: Codestral 2508 with other models
Mistral: Codestral 2508 vs Mistral: Mistral NemoMistral: Codestral 2508 vs Mistral: Mistral Small 3Mistral: Codestral 2508 vs Mistral: Ministral 3 8B 2512 (batch)Mistral: Codestral 2508 vs Mistral: Ministral 3 3B 2512Mistral: Codestral 2508 vs Mistral: Ministral 3 8B 2512Mistral: Codestral 2508 vs Mistral: Mistral Small 3.2 24B
Frequently asked questions
How much does Mistral: Codestral 2508 cost?
Mistral: Codestral 2508 costs $0.3 per 1M input tokens and $0.9 per 1M output tokens (list price, snapshot October 22, 2026). At 10M input + 2M output tokens per month that is about $4.8.
What is the context window of Mistral: Codestral 2508?
Mistral: Codestral 2508 has a context window of 256,000 tokens and can return up to 204,800 output tokens per response.
What are cheaper alternatives to Mistral: Codestral 2508?
Cheaper models include Qwen: Qwen3 Next 80B A3B Instruct ($0.09 in / $1.1 out), Qwen: Qwen3 Coder Flash ($0.195 in / $0.975 out), DeepSeek: DeepSeek V4 Pro 0423 ($0.3835 in / $0.767 out). See the full list on this page.
Is Mistral: Codestral 2508 free to use?
No. Mistral: Codestral 2508 is a paid model at $0.3 input / $0.9 output per 1M tokens. 21 free models are listed on the free models page.
What is the official price of Mistral: Codestral 2508?
On mistralai's official pricing page (checked 2026-10-21), Mistral: Codestral 2508 is listed at $0.3 per 1M input tokens and $0.9 per 1M output tokens. The catalog price tracked on this page matches it exactly.
Provider: mistralai · Model ID: mistralai/codestral-2508 · Modality: text+file->text · Price snapshot October 22, 2026, refreshed daily. How we collect prices