##Build with open-weight models We release open-weight models for everyone to customize and deploy where they want it. Our super-efficient model Mistral Nemo is available under Apache 2.0, while Mistral Large 2 is available through both a free non-commercial license, and a commercial license.
Use them on La Plateforme at market-leading availability, speed, and quality control
Mistral AI is an innovative company specializing in open-source generative AI models, aiming to provide developers and businesses with advanced, customizable AI solutions. Their flagship model, Mistral 7B, is a 7-billion-parameter language model that outperforms larger counterparts like Llama 2 13B across various benchmarks, including reasoning, mathematics, and code generation. This efficiency is achieved through techniques such as grouped-query attention (GQA) and sliding window attention (SWA), enabling faster inference and the handling of longer sequences with reduced computational costs.
In addition to Mistral 7B, the company offers Mixtral, an ensemble of eight Mistral 7B models, providing enhanced performance and a 32K context capacity. These models are proficient in multiple languages, including English, French, Italian, German, Spanish, and various coding languages, making them versatile tools for a wide range of applications.
