Mistral Small 3 is an open-source language model introduced by Mistral AI, featuring 24 billion parameters and operating under the Apache 2.0 license. This model is specifically engineered for low latency and efficient performance, making it suitable for generative AI tasks that require rapid responses. It achieves an accuracy rate of 81% on the Multi-Task Language Understanding (MMLU) benchmark and can generate text at a speed of 150 tokens per second. Mistral Small 3 aims to provide a powerful foundational model for local deployment and customizable development across various industry applications, such as financial services, healthcare, and robotics. The model has not been trained using reinforcement learning (RL) or synthetic data, placing it in the early stages of the production pipeline and making it suitable for building inference capabilities.