Introducing LFM3: Every Modality for Specialized Intelligence

Liquid AI announced LFM3, a new generation of multimodal foundation models in the 2B parameter range designed for specialized intelligence across text, vision, audio, and decision-making tasks. Built with a hybrid architecture combining gated convolutions, sliding-window attention, and global attention, LFM3 maintains efficiency while improving expressiveness and long-context processing compared to its predecessor LFM2. The open-weight models are available on Hugging Face and can run locally on various devices, making them suitable for on-device fine-tuning and deployment.