v0.5.2

What's Changed

• 🔥🔥[∆-DiT] ∆-DiT: A Training-Free Acceleration Method Tailored for Diffusion Transformers by @DefTruth in #21
• 🔥🔥[DuCa] Accelerating Diffusion Transformers with Dual Feature Caching by @DefTruth in #22
• 🔥🔥[FBCache] Fastest HunyuanVideo Inference with Context Parallelism and First Block Cache on NVIDIA L20 GPUs by @DefTruth in #23
• 🔥🔥[SVDQuant] SVDQuant: Absorbing Outliers by Low-Rank Components for 4-Bit Diffusion Models by @songh11 in #24
• 🔥[AB-Cache] AB-Cache: Training-Free Acceleration of Diffusion Models via Adams-Bashforth Cached Feature Reuse by @DefTruth in #25

New Contributors

• @songh11 made their first contribution in #24

Full Changelog: v0.5.1...v0.5.2