Process massive datasets with Devstral 2 2512, featuring an expansive 262K context window for long-document analysis. This model delivers cost-effective pricing at $0.61/1M input and $3.03/1M output tokens, native tool calling support, open weights architecture. Access Devstral 2 2512 via the AI/ML API API with up to 262K output tokens.
Tokens
Tokens
Tokens
Devstral 2 2512 by AI/ML API costs $0.61 per 1M input tokens and $3.03 per 1M output tokens. Cached reads cost $0.06 per 1M tokens.