Codestral 2508
by AI/ML API · family codestral · listed Jul 2025T
Mistral code model for low-latency completion, fill-in-the-middle, and code generation
Current rates USD per 1M tokens
Input$0.413
Output$1.24
Cache read$0.0413
Cache write—
Against the market
On the record
- Model id
mistralai/codestral-2508- Context window
- 256K tokens
- Max output
- 8K tokens
- Input modalities
- text
- Output modalities
- text
- Reasoning
- no
- Tool calls
- yes
- Open weights
- no
- Knowledge cutoff
- —
- Released
- 2025-07-30
- Last updated
- 2025-07-30
- Provider docs
- docs.aimlapi.com
Back of the envelope
≈
Also listed at
- NanoGPT$0.30 in · $0.90 out
- LLM Gateway$0.30 in · $0.90 out
- Kilo Gateway$0.30 in · $0.90 out
- OpenRouter$0.30 in · $0.90 out
- Eden AI$0.30 in · $0.90 out
- Mistral$0.30 in · $0.90 out
- DevPass (LLM Gateway)$0.30 in · $0.90 out
- LowRouter$0.314 in · $0.941 out
- Cortecs$0.368 in · $1.10 out