Nvidia has announced a new partnership with large-scale language model developer Mistral AI, with both companies set to accelerate the development and deployment of a new family of open source models.
Under this partnership, both companies will use Nvidia’s platform to Mr. Mistral Recently announced family, Mistral 3.
Mistral describes this new product range as open source, multilingual, and multimodal with a family optimized across Nvidia’s supercomputing and edge platforms.
Mistral 3 is built using a Mixed Expertise (MoE) architecture. That is, only the relevant parts of the model are activated for the task. This model is touted by companies as enabling more efficient and accurate deployment.
Nvidia said that by combining its GB200 NVL72 system with Mistral AI’s MoE architecture, enterprises will be able to benefit from “high degrees of parallelism” and hardware optimization to more efficiently deploy and scale large-scale models.
“With 41 billion active parameters, 67.5 billion total parameters, and a large 256K context window, Mistral Large 3 provides scalability, efficiency, and adaptability for enterprise AI workloads,” Nvidia said. blog post.
The companies said that as of Dec. 2, the new model is available everywhere “from the cloud to the data center to the edge.”
The partnership builds on existing efforts between the two companies, including the development of the Mistral NeMo 12B language model for chatbots and coding tasks.
Alongside this announcement, Mistral also released nine small language models to help developers run AI “anywhere.”
These models are designed to run on Nvidia hardware, including Spark, RTX PCs, laptops, and Jetson devices. Developers can access models through the AI frameworks Llama.cpp and Ollama.
The Mistral 3 family is available to researchers and developers, which the company said will help “democratize” access to frontier-class AI.
The news will arrive in the same week Nvidia announcement The company announced that it has invested $2 billion in chipmaker Synopsys, underscoring the company’s efforts to strengthen its position in the AI and computing space.
