MultimodalOpen weightsFree
SenseNova-U
SenseTime
SenseNova-U is an image & video generation model line from SenseTime. The latest release is SenseNova-U1.5 (19 Aug 2026). Specs, licensing, and availability below. Official site →
SenseNova-U1.5
Open weightsFreeCommunity APISelf-host
Unified understanding + generation checkpoint built on NEO-unify; better text and infographic rendering, native 4K, image editing with multi-reference, region/box control
- License
- Apache 2.0
- Commercial use
- Allowed
- Availability
- Self-host, Community API
- Parameters
- 8B MoT
- Resolution
- 4K native
Dated to the Hugging Face repo creation (2026-08-19); no SenseTime announcement page located.
See SenseNova-U in context on the image model timeline, or browse all model families.