Codestral Embed is a language model from Mistral AI. It features an 8K context window, code generation, function calling. It is released under the Apache 2.0 license.
Input
Output
Context
8K
Max Output
8K
Parameters
—
Input Modalities
Output Modalities
Features
Data sourced from official provider APIs and documentation
Last updated: May 5, 2026
From model selection to production, one platform, no fragmentation.