Multimodal model

VisionPsy Nano 460M Flash

VisionPsy-Nano-460M-Flash is the efficiency-optimized sibling of VisionPsy-Nano-460M, sharing the same ~460M parameter nanoVLM architecture but processing images with far fewer visual tokens (as few as 64 versus 256 to 1088 for peers). This gives it roughly an order of magnitude faster time-to-first-token and a smaller memory footprint on phones, while retaining about 99% of the full model's benchmark quality.