OrbitQuant: Data-Agnostic Quantization for Image and Video Diffusion Transformers
Diffusion transformers (DiTs) achieve state-of-the-art image and video generation, but their multi-step sampling and growing parameter count make inference expensive. Post-training quantization (PTQ) is the natural remedy, yet DiT activations shift across timesteps, prompts, and guidance branches, forcing prior methods to re-fit calibration data for every new checkpoint or modality. We present OrbitQuant, a data-agnostic weight-activation quantizer that bypasses range estimation by quantizing in a normalized, rotated basis. In this basis, a randomized permuted block-Hadamard (RPBH) rotation co
Lineage graph
Paper → model → repo connections mined from source citations (Tier-1 exact match).
Why these links exist
- Linked via arxiv authorDonghyun Lee →
OrbitQuant: Data-Agnostic Quantization for Image and Video Diffusion Transformers
- Linked via arxiv authorJitesh Chavan →
OrbitQuant: Data-Agnostic Quantization for Image and Video Diffusion Transformers
- Linked via arxiv authorDuy Nguyen →
OrbitQuant: Data-Agnostic Quantization for Image and Video Diffusion Transformers
- Linked via arxiv authorSam Huang →
OrbitQuant: Data-Agnostic Quantization for Image and Video Diffusion Transformers
- Linked via arxiv authorLiming Jiang →
OrbitQuant: Data-Agnostic Quantization for Image and Video Diffusion Transformers
- Linked via arxiv authorPriyadarshini Panda →
OrbitQuant: Data-Agnostic Quantization for Image and Video Diffusion Transformers
- Linked via arxiv authorTimo Mertens →
OrbitQuant: Data-Agnostic Quantization for Image and Video Diffusion Transformers
- Linked via arxiv authorSaurabh Shukla →
OrbitQuant: Data-Agnostic Quantization for Image and Video Diffusion Transformers