Liquid AI Launches Open-Weight d1 Models Featuring Zero Output Tokens
Liquid AI has released the open-weight d1-3B and d1-omni-600M multimodal decision models, which process text, image, and audio inputs to deliver typed, calibrated answers in a single forward pass without generating output tokens. This architecture aims to enable ultra-low-latency, real-time decision-making for edge devices and automated systems.
Source: MarkTechPost