Building on our existing IBM watsonx.ai catalogue of Mistral models, Mistral Medium 3 is now available. Built to meet the enterprise need for high accuracy results coupled with cost-effective scalability, Mistral Medium delivers. Its efficient design allows for deployment on just 4xH100 GPUs, making it an ideal solution for on-premise organizations seeking powerful—yet manageable—AI infrastructure.

About Mistral Medium

Mistral Medium 3 (25.05) features multimodal capabilities and an extended context length of up to 128k. It can now process and understand both visual inputs as well as long documents, further expanding its range of applications. This versatile model has been designed for various tasks including......

Click here to read the full blog.