SeedEdit 3.0: Fast and High-Quality Generative Image Editing

TL;DR

SeedEdit 3.0 is a significantly improved image editing model that combines advanced data curation, joint learning, and a new T2I model to achieve better instruction following and content preservation on real images.

Contribution

The paper introduces SeedEdit 3.0 with enhanced data strategies and joint learning, advancing high-quality, fast generative image editing capabilities.

Findings

Achieves a 56.1% usability rate on benchmarks.

Outperforms previous versions and competitors in image editing quality.

Effective scaling of editing data through new curation pipeline.

Abstract

We introduce SeedEdit 3.0, in companion with our T2I model Seedream 3.0, which significantly improves over our previous SeedEdit versions in both aspects of edit instruction following and image content (e.g., ID/IP) preservation on real image inputs. Additional to model upgrading with T2I, in this report, we present several key improvements. First, we develop an enhanced data curation pipeline with a meta-info paradigm and meta-info embedding strategy that help mix images from multiple data sources. This allows us to scale editing data effectively, and meta information is helpfult to connect VLM with diffusion model more closely. Second, we introduce a joint learning pipeline for computing a diffusion loss and reward losses. Finally, we evaluate SeedEdit 3.0 on our testing benchmarks, for real/synthetic image editing, where it achieves a best trade-off between multiple aspects, yielding a high usability rate of 56.1%, compared to SeedEdit 1.6 (38.4%), GPT4o (37.1%) and Gemini 2.0 (30.3%).