title: "Un Ministral, dei Ministraux | Mistral AI"

Mistral AI has introduced two new state-of-the-art models, Ministral 3B and Ministral 8B, designed for on-device and edge computing applications. These models establish new performance benchmarks in the sub-10B category for knowledge, reasoning, and function-calling, outperforming competitors like Gemma and Llama. Key use cases include on-device translation, offline smart assistants, local analytics, and serving as efficient intermediaries in multi-step agentic workflows. Both models support a context length of up to 128k tokens, with Ministral 8B featuring a special sliding-window attention pattern for faster, memory-efficient inference.