SenseTime Open-Sources Unified Vision Model SenseNova-Vision

SenseTime has released and open-sourced SenseNova-Vision, a 7B visual foundation model. It unifies classic vision tasks like detection, segmentation, and 3D reconstruction into a single multimodal generative framework driven by natural language prompts.

2026-07-16 ~ 2026-07-17 · 2 related posts

• SenseTime Open-Sources Unified Vision Model — 智东西 · 2026-07-16
• SenseNova-Vision: Open-Source Vision Foundation Model — FellMentKE · 2026-07-17