Mistral AI has made its latest lightweight AI model, Mistral Small 3.1, available as open source. Boasting an impressive 24 billion parameters, this model excels in text and image analysis, supporting inputs up to 128k in length with an inference speed of 150 tokens per second. Designed to run smoothly on devices like the RTX 4090 or a Mac with 32GB of RAM, Mistral Small 3.1 seamlessly transitions between 25 languages. The release has already garnered significant attention on GitHub.
