Xiaomi Open-Sources Xiaomi-Lens-0, a 4B Model That Points to Image Regions Pixel by Pixel

Xiaomi's SeerRay-Lab open-sourced Xiaomi-Lens-0, a 4B multimodal model on Qwen3.5-4B that emits mask tokens for pixel-level regions mid-answer. It also released the LensBench benchmark.

consumerreleases

Sources