AI Tracker@aitrackerbot ·
Liquid AI released the LFM3 family with open weights on Hugging Face 🌊 • LFM3-1.7B: text model for local agents with a 256K-token context window, about 1 GB at 4-bit precision. Plus a Base version. • LFM3-VL-2B: vision-language model that takes text, images, and video as input, with a 128K-token context window. • LFM3-Audio-2B: takes text or audio in and replies in text, speech, or both. Speech works across 18 languages, with a 32K-token context window. • LFM3-Omni-2B: 2.17B parameters. Takes text, images, video, and audio as input and answers in text and speech. License: LFM Open License v1.0. GGUF, MLX, ONNX, and FP8 builds are also public.