SenseTime Releases SenseNova-U1.5, an 8B Encoder-Free Unified Multimodal Model
SenseTime has released SenseNova-U1.5, an 8B unified multimodal model that performs visual understanding, reasoning and image generation without external encoders or VAEs. The model supports 4K resolution and is now available on Hugging Face.
2026-09-11 ~ 2026-09-11 · 2 related posts
- SenseNova-U1.5: an 8B encoder-free native unified multimodal model for understanding and generation — Haiwen Diao · 2026-09-11
- SenseTime's SenseNova-U1.5 hits Hugging Face: 8B encoder-free multimodal model with native 4K — liuziwei7 · 2026-09-11