Mistral Large 3

The Mistral Large 3 family from Mistral AI delivers a state-of-the-art multimodal MoE model built for high-performance reasoning and long-context tasks. Featuring 675B total parameters and 41B active, it enables efficient, production-grade deployment for advanced agentic, enterprise, and multimodal applications.

1 variant

Mistral Large 3 675B Instruct 2512

Mistral

Params: 675B

Mistral Large 3 675B Instruct 2512 is a large-scale multimodal Mixture-of-Experts (MoE) model designed for instruction following, agentic workflows, and long-context understanding. It has 675B total parameters with about 41B activated, using a 61-layer transformer with 128 attention heads and a 7,168 hidden size. The architecture includes 128 experts with 4 selected per token and supports a context window of ~294K tokens. It also integrates a 2.5B vision encoder and FP8 quantization, enabling efficient deployment and strong performance across text and vision tasks.

Text GenerationInstruction Following+8

Runs onB200+

Runs on

Mistral Large 3 Models | Vultr Docs