Emu3.5: Native Multimodal Models are World Learners

TL;DR

Emu3.5 is a large-scale multimodal world model trained on extensive vision-language data, capable of complex generation, reasoning, and world modeling, with significant inference speed improvements and open-source availability.

Contribution

The paper introduces Emu3.5, a novel unified multimodal model with efficient inference, advanced capabilities, and extensive training on interleaved vision-language data, advancing multimodal AI research.

Findings

Achieves strong multimodal generation and reasoning performance.

Accelerates inference by approximately 20x with DiDA.

Demonstrates generalizable world-modeling and open-world manipulation.

Abstract

We introduce Emu3.5, a large-scale multimodal world model that natively predicts the next state across vision and language. Emu3.5 is pre-trained end-to-end with a unified next-token prediction objective on a corpus of vision-language interleaved data containing over 10 trillion tokens, primarily derived from sequential frames and transcripts of internet videos. The model naturally accepts interleaved vision-language inputs and generates interleaved vision-language outputs. Emu3.5 is further post-trained with large-scale reinforcement learning to enhance multimodal reasoning and generation. To improve inference efficiency, we propose Discrete Diffusion Adaptation (DiDA), which converts token-by-token decoding into bidirectional parallel prediction, accelerating per-image inference by about 20x without sacrificing performance. Emu3.5 exhibits strong native multimodal capabilities, including long-horizon vision-language generation, any-to-image (X2I) generation, and complex text-rich image generation. It also exhibits generalizable world-modeling abilities, enabling spatiotemporally consistent world exploration and open-world embodied manipulation across diverse scenarios and tasks. For comparison, Emu3.5 achieves performance comparable to Gemini 2.5 Flash Image (Nano Banana) on image generation and editing tasks and demonstrates superior results on a suite of interleaved generation tasks. We open-source Emu3.5 at https://github.com/baaivision/Emu3.5 to support community research.