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

Liquid AI has introduced the d1-3B and d1-omni-600M models, which process text, image, and audio inputs to provide calibrated, typed outputs. Unlike standard generative models, these systems operate with zero output tokens by generating specific answers in a single forward pass. This architecture is designed for decision-making tasks where efficiency is prioritized over generating natural language text.