diff --git a/CHANGELOG.md b/CHANGELOG.md index b37ca34cc..fa13aeacb 100644 --- a/CHANGELOG.md +++ b/CHANGELOG.md @@ -71,6 +71,7 @@ Although upgrades and existing installations are tested and should work fine! with *t2i, i2i, v2v* workflows optional native upsampling and video refine workflows LoRA support with different conditioning types such as Canny/Depth/Pose, etc. + - support for post load quantization - **UI** - major update to modernui layout - redesign of the *Flat* UI theme @@ -105,6 +106,9 @@ Although upgrades and existing installations are tested and should work fine! - **Batch** warn on unprocessable images and skip operations on errors so that other images can still be processed - **Metadata** improved parsing and detect foreign metadata detect ComfyUI images +- **SDNQ** + - use inference context during quantization + - use static compile - **API** - add `/sdapi/v1/lock-checkpoint` endpoint that can be used to lock/unlock model changes if model is locked, it cannot be changed using normal load or unload methods @@ -122,9 +126,9 @@ Although upgrades and existing installations are tested and should work fine! - fix xyz with control enabled - fix control order of image save operations - fix control batch-input processing + - fix torchvision bicubic upsample with ipex - cleanup control infotext - allow upscaling with models that have implicit VAE processing - - sdnq use inference context during quantization - framepack improve offloading - improve scripts error handling - improve infotext param parsing