Liquid AI Unveils LFM2.5-VL-3B Vision-Language Model Optimized for Edge Devices

Liquid AI has introduced LFM2.5-VL-3B, a 3.1 billion parameter vision-language model tailored for on-device applications such as smartphones and laptops. The model features enhanced capabilities in screen understanding, object grounding, multi-image reasoning, and function calling, demonstrating superior performance on real-world image and digital content tasks compared to larger models. It combines a SigLIP2 vision encoder with a large text model backbone and is trained on extensive multimodal,