Abstract
Diffusion transformers (DiTs) achieve state-of-the-art image generation, but their sampling cost limits deployment. Quantizing both weights and activations to 4 bits reduces this cost, yet existing methods fall short in one of two ways. Calibration-based methods are tied to a specific checkpoint and prompt distribution, whereas data-free Hadamard rotation, effective for LLMs, loses quality on DiTs. We show that this loss has a structural cause. Adaptive layer-norm conditioning adds a per-token mean to the activations, and at the widths of the evaluated DiTs, the Hadamard rotations used by data-free methods cannot spread this mean uniformly across coordinates. A single dominant direction therefore survives the rotation and sets the quantization range. We introduce CentriQ, a calibration-free quantizer that centers each token before rotation and restores the mean exactly through a rank-1 full-precision branch, so that per-token scales follow in closed form without data. Weights are fitted under a robust $\ell_p$ objective that tracks the dense mode of each group and discounts heavy tails. Across three DiTs, CentriQ matches the quality of calibrated SVDQuant at 4 bits, whereas calibration-free weight quantizers with plain per-token activation quantization collapse or degrade substantially. CentriQ outperforms the strongest calibration-free method reported to date at 2-bit weights. It is also the first calibration-free method to retain usable image quality at 2-bit activations.
Keywords
Subject
Publication details
- Journal
- Not available
- Open access
- Green open access
Cite this article
APA 7
Jovanović, N., Salzmann, M., & Javed, S. (2026). CentriQ: Calibration-Free Quantization of Diffusion Transformers via Exact Mean Centering. https://omanscience.com/en/articles/centriq-calibration-free-quantization-of-diffusion-transformers-via-exact-mean-centering
MLA 9
Jovanović, Nataša, et al. "CentriQ: Calibration-Free Quantization of Diffusion Transformers via Exact Mean Centering." https://omanscience.com/en/articles/centriq-calibration-free-quantization-of-diffusion-transformers-via-exact-mean-centering.
Chicago (author–date)
Jovanović, Nataša, Mathieu Salzmann, and Saqib Javed. 2026. "CentriQ: Calibration-Free Quantization of Diffusion Transformers via Exact Mean Centering." https://omanscience.com/en/articles/centriq-calibration-free-quantization-of-diffusion-transformers-via-exact-mean-centering.
Harvard
Jovanović, N., Salzmann, M. and Javed, S. (2026) 'CentriQ: Calibration-Free Quantization of Diffusion Transformers via Exact Mean Centering', Available at: https://omanscience.com/en/articles/centriq-calibration-free-quantization-of-diffusion-transformers-via-exact-mean-centering.
Vancouver
Jovanović N, Salzmann M, Javed S. CentriQ: Calibration-Free Quantization of Diffusion Transformers via Exact Mean Centering. https://omanscience.com/en/articles/centriq-calibration-free-quantization-of-diffusion-transformers-via-exact-mean-centering
IEEE
N. Jovanović, M. Salzmann, and S. Javed, "CentriQ: Calibration-Free Quantization of Diffusion Transformers via Exact Mean Centering," https://omanscience.com/en/articles/centriq-calibration-free-quantization-of-diffusion-transformers-via-exact-mean-centering.