GigaTok: Scaling Visual Tokenizers to 3 Billion Parameters for Autoregressive Image Generation

Jiashi Feng, Xihui Liu, Zilong Huang, Jun Liew, Tianwei Xiong

arXiv · PDF · Open in the Atlas

Code that ran

We lifted 15 functions out of this paper's own repositories and ran 10 of them in a sandbox. "Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results.

Repositories linked to this paper

• SilentView/GigaTok — marked official in Papers with Code

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

In autoregressive (AR) image generation, visual tokenizers compress images into compact discrete latent tokens, enabling efficient training of downstream autoregressive models for visual generation via next-token prediction. While scaling visual tokenizers improves image reconstruction quality, it often degrades downstream generation quality-a challenge not adequately addressed in existing literature. To address this, we introduce GigaTok, the first approach to simultaneously improve image reconstruction, generation, and representation learning when scaling visual tokenizers. We identify the growing complexity of latent space as the key factor behind the reconstruction vs. generation dilemma. To mitigate this, we propose semantic regularization, which aligns tokenizer features with semantically consistent features from a pre-trained visual encoder. This constraint prevents excessive latent space complexity during scaling, yielding consistent improvements in both reconstruction and downstream autoregressive generation. Building on semantic regularization, we explore three key practices for scaling tokenizers: (1) using 1D tokenizers for better scalability, (2) prioritizing decoder scaling when expanding both encoder and decoder, and (3) employing entropy loss to stabilize training for billion-scale tokenizers. By scaling to 3 billion parameters, GigaTok achieves stateof-the-art performance in reconstruction, downstream AR generation, and downstream AR representation quality.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2504.08736")
get_code_for_paper("2504.08736")
have("2504.08736")

Connect an agent — have() is free.