# TreeQ: Pushing the Quantization Boundary of Diffusion Transformer via Tree-Structured Mixed-Precision Search Source: [TreeQ: Pushing the Quantization Boundary of Diffusion Transformer via Tree-Structured Mixed-Precision Search]() Collection index: [llms.txt]() Data updated: 2025-12-10T05:04:31.849Z Authors: Kaicheng Yang, Kaisen Yang, Baiting Wu, Xun Zhang, Qianrui Yang, Haotong Qin, He Zhang, Yulun Zhang Published: 2025-12-06T08:59:12.000Z ## Abstract Diffusion Transformers (DiTs) have emerged as a highly scalable and effective backbone for image generation, outperforming U-Net architectures in both scalability and performance. However, their real-world deployment remains challenging due to high computational and memory demands. Mixed-Precision Quantization (MPQ), designed to push the limits of quantization, has demonstrated remarkable success in advancing U-Net quantization to sub-4bit settings while significantly reducing computational and memory overhead. Nevertheless, its application to DiT architectures remains limited and underexplored. In this work, we propose TreeQ, a unified framework addressing key challenges in DiT quantization. First, to tackle inefficient search and proxy misalignment, we introduce Tree Structured Search (TSS). This DiT-specific approach leverages the architecture's linear properties to traverse the solution space in O(n) time while improving objective accuracy through comparison-based pruning. Second, to unify optimization objectives, we propose Environmental Noise Guidance (ENG), which aligns Post-Training Quantization (PTQ) and Quantization-Aware Training (QAT) configurations using a single hyperparameter. Third, to mitigate information bottlenecks in ultra-low-bit regimes, we design the General Monarch Branch (GMB). This structured sparse branch prevents irreversible information loss, enabling finer detail generation. Through extensive experiments, our TreeQ framework demonstrates state-of-the-art performance on DiT-XL/2 under W3A3 and W4A4 PTQ/PEFT settings. Notably, our work is the first to achieve near-lossless 4-bit PTQ performance on DiT models. The code and models will be available at https://github.com/racoonykc/TreeQ ## Recorded benchmark results A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage. - [Conditional Image Generation on ImageNet-1K 256x256 (val, CFG=1.5)](): Conditional Image Generation; ImageNet 256x256 CFG=1.5 1K (val); IS: 219.66. - [Conditional Image Generation on ImageNet 1K 256x256 CFG=4.0 (val)](): Conditional Image Generation; ImageNet 1K 256x256 CFG=4.0 (val); Inception Score (IS): 470.1. ## Original sources and implementation [Original paper]() [PDF]()