Latest

Zhipu Wujie·Emu3.5 Released, Launching “Next-State Prediction”! Wang Zhongyuan: Could Open the Third Scaling Paradigm

Emu3.5

Zhipu Wujie·Emu3.5 Released, Launching “Next-State Prediction”! Wang Zhongyuan: Could Open the Third Scaling Paradigm

WuJie·Emu3.5 — The Next Leap in Multimodal World Models Introduction In October 2024, the Beijing Academy of Artificial Intelligence (BAAI) released the world’s first natively multimodal world model — WuJie·Emu3. This groundbreaking model is based entirely on next-token prediction, avoiding diffusion or composite methods, and achieves a unified

By Honghao Wang
Shanghai AI Lab Releases Hybrid Diffusion Language Model SDAR: First Open-Source Diffusion Language Model to Surpass 6,600 TGS

SDAR

Shanghai AI Lab Releases Hybrid Diffusion Language Model SDAR: First Open-Source Diffusion Language Model to Surpass 6,600 TGS

Large Model Inference Speed & Cost Bottlenecks SDAR Paradigm as a Breakthrough Large model inference has become slow and costly, creating a core bottleneck that limits broader adoption. The main culprit is the autoregressive (AR) “word-by-word” serial generation paradigm. --- Introduction to SDAR Shanghai Artificial Intelligence Laboratory recently proposed SDAR

By Honghao Wang
Hands-on Test of Meituan’s First Video Foundation Model: Native 5-Minute Photorealistic Long Video Output

AI video

Hands-on Test of Meituan’s First Video Foundation Model: Native 5-Minute Photorealistic Long Video Output

Meituan Launches LongCat-Video — Their First AI Video Model On Monday, Meituan unveiled its first AI video model, LongCat-Video. With 13.6B parameters, a single model can handle: * Text-to-video * Image-to-video * Video continuation * Ultra-long video generation Output: 720p, 30fps. Since my own hardware couldn’t handle full-scale testing, I reached out to

By Honghao Wang