ViDiT-Q: Efficient and Accurate Quantization of Diffusion Transformers for Image and Video Generation

Quantization Method specialized for DiT Figure 1: We introduce ViDiT-Q, a quantization method specialized for diffusion transformers used in image and video generation. ViDiT-Q achieves lossless W8A8 quantization and minimal visual quality degradation at W4A8, gaining 2.5x model size reduction and a 1.5x latency speedup. 1