BITA Conference Releases Wu World Series Large Models to Accelerate the Arrival of the Physical AGI Era

AIbase基地

Published inAI News · 3 min read · Jun 6, 2025

At the Beijing Zhiyuan Conference 2025 held today, the Beijing Zhiyuan Artificial Intelligence Research Institute officially released the "WuJie" series of large models, comprehensively showcasing its latest research results and strategic layout in the direction of physical artificial general intelligence (AGI).

The "WuJie" series aims to break through the boundary between virtual and real worlds and empower the physical world. It covers four cutting-edge models: the multimodal world model Emu3, the neuroscience model JianWei Brainμ, the embodied intelligent brain RoboBrain 2.0, and the microscopic life model OpenComplex2, constructing a full-chain AI system from world understanding, neural modeling to embodied control and life simulation.

Metaverse Science Fiction Cyberpunk Painting (3) Large Model

Among them, Emu3 is an original multimodal generative model that unifies text, image, and video understanding and generation in a self-regressive manner without relying on diffusion architectures, providing modality-agnostic unified representation capabilities. Its core innovation lies in encoding multimodal data into homogeneous token sequences, possessing powerful cross-modal fusion capabilities.

Based on the Emu3 architecture, JianWei Brainμ has achieved tokenization and multimodal alignment of neuroscience signals (such as fMRI, EEG, two-photon imaging) for the first time. The model has completed pre-training on over 1 million neural signal units and is hailed as the "AlphaFold" of the neuroscience field, with broad potential in basic neuroscience research, brain disease diagnosis and treatment, brain-computer interfaces, and more.

Currently, Zhiyuan has collaborated with Peking University, Tsinghua University, Fudan University, the Beijing Institute of Life Sciences, and BrainCo to promote the landing of the "WuJie" models in scientific research and industry, further consolidating China's global competitiveness in the physical AGI path.

BAAI Launches Emu3 and Other WuJie Series Large Models

At the recently concluded 7th Beijing BDA Conference, BAAI released a series of large models named 'WuJie'. The release of these models marks another technological leap in the field of artificial intelligence, drawing the attention of many tech enthusiasts. The models released this time include the native multimodal world model Emu3, brain science multimodal universal foundational model JianWei Brainμ, and the ontology-crossing embodied brain-cerebellum collaboration framework RoboOS2.0 as well as the embodied brain RoboBrain2.0. Additionally, there are full...

BAAI Open-sources Lightweight Ultra-long Video Understanding Model Video-XL-2

Recently, the Beijing Academy of Artificial Intelligence (BAEI), in collaboration with Shanghai Jiao Tong University and other institutions, officially released a new generation of ultra-long video understanding model - Video-XL-2. The introduction of this model represents a significant breakthrough in the field of open-source ultra-long video understanding technology, injecting new vitality into the development of multimodal large models for understanding long video content. In terms of technical architecture, Video-XL-2 mainly consists of three core components: a visual encoder, the Dynamic Token Synthesis (DTS) module, and a large language model (LLM). The model adopts

New Breakthrough in Multimodal Retrieval! Beijing Academy of Artificial Intelligence Open-Sources Multimodal Vector Model BGE-VL

On March 6th, the Beijing Academy of Artificial Intelligence (BAAI) announced the open-sourcing of its multimodal vector model, BGE-VL. This achievement marks a significant breakthrough in the field of multimodal retrieval. The BGE-VL model achieves state-of-the-art results on various multimodal retrieval tasks, including image-text retrieval and compositional image retrieval, significantly improving the performance of multimodal retrieval.

Redefining Multimodal AI! Zhiyuan Releases the Native Multimodal World Model Emu3

Beijing Zhiyuan Artificial Intelligence Research Institute announces the launch of the native multimodal world model Emu3. This model is based on next-token prediction technology and does not rely on diffusion models or combinatorial methods to achieve understanding and generation across text, image, and video modalities. Emu3 surpasses existing well-known open-source models such as SDXL, LLaVA, and OpenSora in tasks like image generation, video generation, and visual language understanding, showcasing exceptional performance.

Beijing Academy of Artificial Intelligence Releases Chinese Internet Corpus CCI3.0 Containing 1000GB Dataset

At the 2024 Beijing Cultural Forum, the Beijing Academy of Artificial Intelligence (BAAI) officially announced the release of the next-generation Chinese Internet corpus CCI3.0 (Chinese Corpora Internet), further promoting data co-construction and sharing. CCI3.0 includes a 1000GB dataset and a 498GB high-quality subset CCI3.0-HQ, marking another important update following the initial open-source release of CCI1.0 in November 2023 and the release of CCI2.0 in April 2024.