Model overview
What this model is for
d1-omni:2b is a compact multimodal decision model from Liquid AI, available through Ollama for local use. The supplied listing identifies image and text as its modalities and associates the model with typed decisions, vision, and audio use cases. This makes it potentially relevant for applications that classify or route typed inputs and analyze images, while audio-related suitability is advertised but not technically detailed in the supplied material. The model page describes it as a 600M decision model, whereas the supplied directory record lists 2 billion parameters, so its effective parameter size should be verified before deployment. No context-window specification, inference-speed benchmark, capabilities list, pricing, licensing terms, or safety classification is available in the supplied material. The page provides an Ollama model reference and installation ecosystem, but it does not document tool calling, coding, agentic behavior, embeddings, image generation, or other specialized features. Developers should validate supported input formats, audio handling, memory requirements, quality on their task, and redistribution or commercial-use permissions before integrating it into production.