Liquid AI has released d1-3B and d1-omni-600M, two open-weight AI models designed for fast, on-device decision-making.
Liquid AI has released d1-3B and d1-omni-600M, two open-weight AI models designed for fast, on-device decision-making.
Unlike traditional generative models, they produce decisions in a single processing pass, supporting real-time analysis of text, images, and audio. d1-3B delivers answers in just 8 milliseconds on an NVIDIA RTX 4090, and both models are available on Hugging Face.