Ming-Omni: A Unified Multimodal Model for Perception and Generation

TL;DR

Ming-Omni is a comprehensive multimodal model that unifies processing and generation of images, text, audio, and video, enabling versatile tasks without multiple specialized models.

Contribution

It introduces a novel unified architecture with modality-specific routers and supports audio and image generation, extending beyond existing multimodal models.

Findings

Achieves strong performance in multimodal perception and generation tasks.

Supports audio and image generation with high quality.

Matches GPT-4o in modality support, setting a new open-source benchmark.

Abstract

We propose Ming-Omni, a unified multimodal model capable of processing images, text, audio, and video, while demonstrating strong proficiency in both speech and image generation. Ming-Omni employs dedicated encoders to extract tokens from different modalities, which are then processed by Ling, an MoE architecture equipped with newly proposed modality-specific routers. This design enables a single model to efficiently process and fuse multimodal inputs within a unified framework, thereby facilitating diverse tasks without requiring separate models, task-specific fine-tuning, or structural redesign. Importantly, Ming-Omni extends beyond conventional multimodal models by supporting audio and image generation. This is achieved through the integration of an advanced audio decoder for natural-sounding speech and Ming-Lite-Uni for high-quality image generation, which also allow the model to engage in context-aware chatting, perform text-to-speech conversion, and conduct versatile image editing. Our experimental results showcase Ming-Omni offers a powerful solution for unified perception and generation across all modalities. Notably, our proposed Ming-Omni is the first open-source model we are aware of to match GPT-4o in modality support, and we release all code and model weights to encourage further research and development in the community.