Mistral AI Releases Mistral Small 4: A 119B-Parameter MoE Model that Unifies Instruct, Reasoning, and Multimodal Workloads

Mistral AI has launched Mistral Small 4, a 119-billion parameter Mixture of Experts (MoE) model that unifies instruction following, reasoning, and multimodal capabilities into a single deployment. This represents the first model from Mistral to consolidate the functions of their previously separate Mistral Small, Magistral, and Pixtral models.

• →Mistral Small 4 is a 119-billion parameter MoE model that combines multiple AI capabilities in one deployment.
• →The model unifies instruction following, reasoning, and multimodal understanding previously handled by separate models.
• →This consolidation eliminates the need for multiple model deployments for different AI workloads.
• →The release represents Mistral AI's strategy to create more versatile and efficient AI models.
• →The model combines capabilities from Mistral Small, Magistral, and Pixtral into a single solution.