U1.5-8B-MoT

#16 v Multimodální modely

Sensenova · 8b · 2× · naposledy 25. 8. 2026

36

Momentum

SenseNova-U1.5-8B-MoT is a native, unified multimodal model from SenseTime/SenseNova that combines image understanding, generation, and editing in a single 8B-parameter architecture (NEO-unify), without a separate Visual Encoder or VAE. It is the current flagship checkpoint of the SenseNova-U1 series and has been improved over its predecessor U1-8B-MoT, particularly in native 4K image generation, text rendering/infographics, and image editing. The weights are Open Source under Apache-2.0 license on Hugging Face and GitHub, including Inference code and training pipeline.

Zdroje (2)