EMMA: Efficient Multimodal Understanding, Generation, and Editing with a Unified Architecture

TL;DR

EMMA is a unified multimodal architecture that combines efficient encoding, reduced token usage, and a shared network to improve understanding, generation, and editing across visual and textual modalities.

Contribution

EMMA introduces a novel architecture with a high compression autoencoder, channel-wise token concatenation, and a shared-decoupled network, advancing efficiency and performance in multimodal tasks.

Findings

EMMA-4B outperforms state-of-the-art models in efficiency and accuracy.

The autoencoder achieves 32x compression, reducing token requirements.

EMMA demonstrates competitive results with recent multimodal models.

Abstract

We propose EMMA, an efficient and unified architecture for multimodal understanding, generation and editing. Specifically, EMMA primarily consists of 1) An efficient autoencoder with a 32x compression ratio, which significantly reduces the number of tokens required for generation. This also ensures the training balance between understanding and generation tasks by applying the same compression ratio to images. 2) Channel-wise concatenation instead of token-wise concatenation among visual understanding and generation tokens, which further reduces the visual tokens in unified architectures. 3) A shared-and-decoupled network that enables mutual improvements across tasks while meeting the task-specific modeling requirements. 4) A mixture-of-experts mechanism adopted for visual understanding encoder, which substantially improves perceptual capabilities with a few parameters increase. Extensive experiments have shown that EMMA-4B can significantly outperform state-of-the-art unified multimodal approaches (e.g., BAGEL-7B) in both efficiency and performance, while also achieving competitive results compared to recent multimodal understanding and generation experts (e.g., Qwen3-VL and Qwen-Image). We believe that EMMA lays a solid foundation for the future development of unified multimodal architectures.