SenseNova-U1-8B-MoT

Haiwen Diao, Penghao Wu, Hanming Deng, Jiahao Wang, Shihao Bai, Silei Wu, Weichen Fan, Wenjie Ye, Wenwen Tong, Xiangyu Fan, Yan Li, Yubo Wang, Zhijie Cao, Zhiqian Lin, Zhitao Yang, Zhongang Cai, Yuwei Niu, Yue Zhu, Bo Liu, Chengguang Lv, Haojia Yu, Haozhe Xie, Hongli Wang, Jianan Fan, Jiaqi Li, Jiefan Lu, Jingcheng Ni, Junxiang Xu, Kaihuan Liang, Lianqiang Shi, Linjun Dai, Linyan Wang, Oscar Qian, Peng Gao, Pengfei Liu, Qingping Sun, Rui Shen, Ruisi Wang, Shengnan Ma, Shuang Yang, Siyi Xie, Siying Li, Tianbo Zhong, Xiangli Kong, Xuanke Shi, Yang Gao, Yongqiang Yao, Yves Wang, Zhengqi Bai, Zhengyu Lin, Zixin Yin, Wenxiu Sun, Ruihao Gong, Quan Wang, Lewei Lu, Lei Yang, Ziwei Liu, Dahua Lin

Open source

Type: model
Venue: SenseTime / SenseNova
Year: 2026
Source: huggingface
Access: free
Language: en
Added: 2026-08-14T18:32:00Z
Verified: 2026-08-14T18:32:00Z

Summary

HF weights for SenseNova-U1-8B-MoT: a dense unified multimodal model (NEO-unify / NEOChat) that drops the visual encoder and VAE and uses Mixture-of-Transformers experts for understanding vs generation. Card: ~8B understanding + ~8B generation params (inspect script: 17.552B total; understanding 8.121B / generation 8.186B / shared embed+lm_head 1.245B). Native interleaved image-text, infographic T2I, editing, VQA; siblings include SFT, Infographic, 8-step preview, LoRA, and A3B-MoT. Code https://github.com/OpenSenseNova/SenseNova-U1. Paper arXiv 2605.12500. Apache-2.0.

Keywords

sensenova-u1 · neo-unify · mot · multimodal · text-to-image · interleaved · sensetime · huggingface

Topics

unified multimodal, image generation, vision-language

Research notes

Primary: HF model card (apache-2.0, transformers, pipeline_tag any-to-any, downloads=28336 likes=291 checked 2026-08-14). Paper arXiv 2605.12500 (cs.CV; Atom API has no license). Official code https://github.com/OpenSenseNova/SenseNova-U1. Collection https://huggingface.co/collections/sensenova/sensenova-u1. Discord posted the HF model pointer. Weights/checkpoint item, not a new hosted pretraining corpus, so no datasets_local row. Unofficial GGUF sibling smthem/SenseNova-U1-8B-MoT-Merger-gguf not copied into hf_* fields.