123 likesNew · 7d old
Omni-modal 2B model runs on phones, answers with voice
What is LFM3-Omni-2B?
LFM3-Omni-2B understands text, images, video and audio, then replies in text or spoken words, all from a single 2.2 billion-parameter model that fits on edge devices.
Who it is for
Developers building mobile or on-device assistants can replace separate speech, vision and language models with one unified engine.