Architecture, strengths and use cases
Devstral Small 2 24B is an official self-hostable Mistral AI release in the Mistral Devstral track. It uses Mistral3ForConditionalGeneration, carries 24B, supports a verified context window of 393,216 tokens, and exposes only source-certified checkpoints and runtime combinations. Its release, license and deployment facts are tied to the official repository revision rather than inferred from the family name.
Devstral Small 2 24B uses Mistral3ForConditionalGeneration with 24B. The stored architecture profile records 40 layers, hidden width 5120, 32 attention heads and a 393,216 token context ceiling where the official configuration publishes those fields.
Verified Mistral3ForConditionalGeneration architecture
393,216 token certified context
Official FP8 checkpoint coverage
Private LLM inference
Agentic workflows
Long-context document processing
Software engineering and code generation