Fan Zhang

Technical Staff @BAAIbeijing

[BAAI @BAAIBeijing]

🚀 Introducing Emu3.5 — a large-scale multimodal world model that natively predicts the next vision-language state.
🔥 Trained on over 10T interleaved vision-language tokens and enhanced with reinforcement learning, Emu3.5 achieves powerful multimodal reasoning and generation.
⚡ Powered by our new Discrete Diffusion Adaptation (DiDA) for 20× faster inference.
🔥 Emu3.5 outperforms Nano Banana across image generation, editing, interleaved tasks and more.

🌍 Explore Emu3.5: https://t.co/pWq9wKNXtm

Github: https://t.co/bsjNkwypTa

#Emu3 #MultimodalAI #WorldModel #NextTokenPrediction

[Fan Zhang @ryanzhangfan]

🚀 Introducing ETT: the 1st end-to-end vision tokenizer tuning approach designed for native multimodal models - jointly optimizing the vision tokenizer and autoregressive LLM for better representation alignment!
✨ Let your tokenizer do more than reconstruct - let it understand!

[Fan Zhang @ryanzhangfan]

🚀We’re excited to launch a demo for our latest work, Emu3 - a SOTA multimodal models trained via only next-token prediction - on @huggingface! Feel free to try it out!

👉demo link: https://t.co/8nUVAWyMMX
🔗project page: https://t.co/StVw6dCkAX
🔍github: https://t.co/OXepTijXvG