mlsanityData to 7 Oct 2026

OrbitQuant: Data-Agnostic Quantization for Image and Video Diffusion Transformers

arXiv 2607.02461 ↗First posted 2 Jul 2026cs.CV cs.AI cs.LGcited by 5

On the map: Diffusion models → Diffusion foundations and sampling

Abstract (arXiv)

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 concentrates each coordinate around one fixed, known marginal regardless of the input, so a single Lloyd-Max codebook serves all timesteps, prompts, and layers of a given input dimension. We extend the same quantizer to weight rows offline, absorbing the rotation into the weights so that it cancels inside each linear layer and only a forward rotation on the activations remains at runtime. The same recipe transfers from image to video with no per-modality tuning. Across FLUX.1, Z-Image-Turbo, Wan 2.1, and CogVideoX, it sets the state of the art for PTQ at several low-bit settings. It also pushes PTQ of image diffusion transformers to W2A4 with usable generation quality.

Key facts. Cited by 5 papers in the map (data through 7 Oct 2026). Placed in Diffusion foundations and sampling, one of 14 directions in Diffusion models. Builds on 43 of its 52 references that are in the map. First posted on arXiv 2 Jul 2026.

Builds on

43 of 52 references.

Closest ancestors · 21

Same direction, newest first

  1. 2026PermuQuant: Lowering Per-Group Quantization Error by Reordering Channels for Diffusion Models10 May
  2. 2026AdaTSQ: Pushing the Pareto Frontier of Diffusion Transformers via Temporal-Sensitivity Quantization10 Feb
  3. 2025ConvRot: Rotation-Based Plug-and-Play 4-bit Quantization for Diffusion Transformers3 Dec
  4. 2025Z-Image: An Efficient Image Generation Foundation Model with Single-Stream Diffusion Transformer27 Nov
  5. 2025RobuQ: Pushing DiTs to W1.58A2 via Robust Activation Quantization28 Sep
  6. 2025S2Q-VDiT: Accurate Quantized Video Diffusion Transformer with Salient Data and Sparse Token Distillation6 Aug
  7. 2025LRQ-DiT: Log-Rotation Post-Training Quantization of Diffusion Transformers for Image and Video Generation5 Aug
  8. 2025DVD-Quant: Data-free Video Diffusion Transformers Quantization24 May
  9. 2025QVGen: Pushing the Limit of Quantized Video Generative Models16 May
  10. 2024SVDQuant: Absorbing Outliers by Low-Rank Components for 4-Bit Diffusion Models7 Nov
  11. 2024SANA: Efficient High-Resolution Image Synthesis with Linear Diffusion Transformers14 Oct
  12. 2024Q-DiT: Accurate Post-Training Quantization for Diffusion Transformers25 Jun
9 more
  1. 2024ViDiT-Q: Efficient and Accurate Quantization of Diffusion Transformers for Image and Video Generation4 Jun
  2. 2024PTQ4DiT: Post-training Quantization for Diffusion Transformers25 May
  3. 2024PixArt-Sigma: Weak-to-Strong Training of Diffusion Transformer for 4K Text-to-Image Generation7 Mar
  4. 2024Scaling Rectified Flow Transformers for High-Resolution Image Synthesis5 Mar
  5. 2023EfficientDM: Efficient Quantization-Aware Fine-Tuning of Low-Bit Diffusion Models5 Oct
  6. 2022Scalable Diffusion Models with Transformers19 Dec
  7. 2022All are Worth Words: A ViT Backbone for Diffusion Models25 Sep
  8. 2021High-Resolution Image Synthesis with Latent Diffusion Models20 Dec
  9. 2020Denoising Diffusion Probabilistic Models19 Jun

Other references · 22

  1. 2022Photorealistic Text-to-Image Diffusion Models with Deep Language Understandingcited by 6,322
  2. 2025Wan: Open and Advanced Large-Scale Video Generative Modelscited by 2,536
  3. 2024CogVideoX: Text-to-Video Diffusion Models with An Expert Transformercited by 2,198
  4. 2022GPTQ: Accurate Post-Training Quantization for Generative Pre-trained Transformerscited by 1,838
  5. 2022Make-A-Video: Text-to-Video Generation without Text-Video Datacited by 1,725
  6. 2024HunyuanVideo: A Systematic Framework For Large Video Generative Modelscited by 1,597
  7. 2022SmoothQuant: Accurate and Efficient Post-Training Quantization for Large Language Modelscited by 1,338
  8. 2023AWQ: Activation-aware Weight Quantization for LLM Compression and Accelerationcited by 1,234
  9. 2025Qwen-Image Technical Reportcited by 973
  10. 2024QuaRot: Outlier-Free 4-Bit Inference in Rotated LLMscited by 598
  11. 2024SpinQuant: LLM quantization with learned rotationscited by 446
  12. 2023QuIP: 2-Bit Quantization of Large Language Models With Guaranteescited by 397
10 more
  1. 2023SpQR: A Sparse-Quantized Representation for Near-Lossless LLM Weight Compressioncited by 359
  2. 2024QuIP#: Even Better LLM Quantization with Hadamard Incoherence and Lattice Codebookscited by 354
  3. 2023SqueezeLLM: Dense-and-Sparse Quantizationcited by 285
  4. 2024DuQuant: Distributing Outliers via Dual Transformation Makes Stronger Quantized LLMscited by 165
  5. 2024FlatQuant: Flatness Matters for LLM Quantizationcited by 116
  6. 2025TurboQuant: Online Vector Quantization with Near-optimal Distortion Ratecited by 116
  7. 2025OstQuant: Refining Large Language Model Quantization with Orthogonal and Scaling Transformations for Better Distribution Fittingcited by 94
  8. 2024Fast Matrix Multiplications for Lookup Table-Quantized LLMscited by 27
  9. 2026Pushing the Limits of Block Rotations in Post-Training Quantizationcited by 9
  10. 2026LoPRo: Enhancing Low-Rank Quantization via Permuted Block-Wise Rotationcited by 3

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2026: 5.

5 papers.

  1. 2026RSLM: Training-Free Vector Quantization for Approximate Nearest Neighbor Searchcited by 0
  2. 2026PulseQuant: Propagation-Guided Subspace Correction for 4-Bit Video Diffusion Transformerscited by 0
  3. 2026Joint Branch-Space Transform Coding for Diffusion Activation Quantization with Classifier-Free Guidancecited by 0
  4. 2026SoloQ: Calibration-Free Quantization for Diffusion Language Modelscited by 0
  5. 2026Deflating the Hessian: Rank-4 W4A4 Quantization for Multimodal Diffusion Transformerscited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2026CentriQ: Calibration-Free Quantization of Diffusion Transformers via Exact Mean Centering
  2. 2026DiRotQ: Rotation-Aware Quantization for 4-bit Diffusion Transformers
  3. 2025TR-DQ: Time-Rotation Diffusion Quantization
  4. 2024TaQ-DiT: Time-aware Quantization for Diffusion Transformers
  5. 2024VQ4DiT: Efficient Post-Training Vector Quantization for Diffusion Transformers
  6. 20266Bit-Diffusion: Inference-Time Mixed-Precision Quantization for Video Diffusion Models
  7. 2024An Analysis on Quantizing Diffusion Transformers
  8. 2024DiTAS: Quantizing Diffusion Transformers via Enhanced Activation Smoothing