Liquid AI Releases Open-Weight d1-3B and d1-omni-600M: Multimodal Decision Models With Zero Output Tokens

Liquid AI has released the d1-3B and d1-omni-600M models, which are designed to process multimodal inputs like text, images, and audio. Unlike standard generative models, these systems function as decision models that provide direct, typed answers in a single forward pass without generating output tokens. This approach aims to improve efficiency and predictability for tasks requiring specific classifications or outputs rather than open-ended text generation.