From 27509a80bb5da799cd007f878990f6f6843b35f7 Mon Sep 17 00:00:00 2001 From: n00mkrad Date: Mon, 30 Jan 2023 23:44:01 +0100 Subject: [PATCH] Changelogs & version file --- changelog-motd.txt | 78 ++++++++++++++-------------------------------- changelog.txt | 18 +++++++++++ meta/version.json | 2 +- 3 files changed, 43 insertions(+), 55 deletions(-) diff --git a/changelog-motd.txt b/changelog-motd.txt index 962221a9..5f436672 100644 --- a/changelog-motd.txt +++ b/changelog-motd.txt @@ -1,3 +1,26 @@ +SD GUI 1.9.1 Changelog: +- New: Resolution can now also be adjusted in InstructPix2Pix mode (resizes input image) +- New: Resulting aspect ratio is shown next to resolution option +- New: Right-click image menu in InstructPix2Pix has an option to copy a comparison image +- Improved: Model compatibility, any Safetensors file should now work fine after conversion +- Improved: Model converter no longer includes 1 GB safety checker in ONNX outputs +- Improved: Image Viewer now shows image resolution info in InstructPix2Pix mode +- Improved: InstructPix2Pix no longer downloads unnecessary files, now <2GB +- Fixed: InstructPix2Pix errors when trying to run it without internet connection +- Fixed: Certain options result in filenames being too long causing issues +- Fixed: Stable Diffusion model gets reloaded if amount of free RAM changed +- Fixed: Image export breaks if initialization image(s) import takes too long +- Fixed: "Open Output Folder" opens Documents folder +- Fixed: ONNX and InstructPix2Pix don't accept images with UTF8 characters in path +- Fixed: Running InvokeAI post-processing sometimes fails due to long paths +- Fixed: Issues with comma vs dot, depending on Windows region settings + +Notes: +- InstructPix2Pix supports resolutions divisible by 8, not only 64 like regular SD +- SD 2.x models are not yet supported, scheduled for next major update + + + SD GUI 1.9.0 Changelog: - New: InstructPix2Pix (Settings -> Image Generation Implementation -> InstructPix2Pix) - New: Option to show the input image next to the output for comparisons @@ -7,13 +30,6 @@ SD GUI 1.9.0 Changelog: - Fixed: Minor PNG metadata parsing issues - Fixed: Various of other minor fixes -Notes: -- InstructPix2Pix will download its model files (2.6 GB) on the first run -- InstructPix2Pix works with any resolution, not only those divisible by 64 -- InstructPix2Pix currently doesn't work with filenames containing special characters -- SD 2.x models are not yet supported, scheduled for next major update -- "Open Output Folder" button currently does not work, use the right-click menu instead - SD GUI 1.8.1 Changelog: @@ -114,50 +130,4 @@ SD GUI 1.7.0 Changelog: - "Delete All Current Images" now requires a confirmation - Improved prompt history/queue UI (show full prompt on hover, and more) - Improved GUI rendering, should have less flickering now -- Fixed bug where model pruning would say "Failed..." even if it was successful - - - -SD GUI 1.6.0 Changelog: -- Added Dreambooth training (24 GB VRAM required for now!) -- Added support for prompt emphasis - () for more, {} for less -- Added Model Quick Switcher: Press Ctrl+M to quickly change the current AI model -- Added model folder manager: You can now add additional model folders to load models from -- Pop-Up Image Viewer: When slideshow mode is enabled, Left/Right arrow keys change images -- Pop-Up Image Viewer: Window can now be resized/zoomed in 25% steps using mouse wheel -- Pop-Up Image Viewer: Added "Always on Top" option to keep window always on top -- Added lots of hotkeys, documented on Github Readme -- Words in prompt field can now be deleted using CTRL+BACKSPACE, like in most text editors -- Model Pruning Tool: Added option to delete input file if pruning was successful -- Fixed an issue where the Stable Diffusion process would be killed when cancelling -- Fixed prompt queue not working after running first entry - - - -SD GUI 1.5.0 Changelog: -- SD Update: Supports exclusion words, runs on 4 GB VRAM (when no other apps are open) -- UI is now more flexible, window can be resized, prompt field is bigger and has zoomable text -- Added CodeFormer face restoration as an alternative to GFPGAN -- Updated RealESRGAN (upscaler), should now be faster with same or better quality -- Added button to delete either the current image, or all generated images -- Added separate checkboxes for including prompt/seed/scale/sampler/model in filename -- Added option to save original image in addition to the post-processed image (if postproc is enabled) -- Added option to select the CUDA device (Automatic, CPU, or a specific GPU) -- Added model merging tool -- Added model pruning tool (strip EMA data and/or convert to fp16 half-precision for 2 GB models) -- Added option to unload Stable Diffusion after each generation (like in pre-1.4.0) -- Added reliable orphan process handling (Python no longer stays in RAM if the GUI crashes) -- Image Viewer slideshow: Added short cooldown after using prev/next image buttons -- Image Viewer context menu: Added button to re-generate single image with current settings/seed -- Image Viewer Pop-up: Now borderless, 100% zoom by default, double-click for fullscreen -- Image Viewer Pop-up: Added Slideshow Mode which mirrors the regular image viewer when enabled -- Added image load form, allows you to use as init image, load settings from metadata, or copy prompt -- Images can now be loaded from clipboard, not just from files -- Prompt History: Added option to disable history, added text filter -- You can now add an entry to the prompt queue by right-clicking on its icon -- Disabled post-processing with Low Memory Mode as it was not working properly -- Prompt text in filenames now strips weighting (won't create new folders for each weight change) -- Current model name gets printed whenever Stable Diffusion is started -- Full Precision is now enabled by default on GTX 16 series cards to fix compatibility with them -- Fixed empty/invalid prompts (e.g. newlines) counting towards the target image amount -- Some fixes regarding cancelling the generation process and handling crashes \ No newline at end of file +- Fixed bug where model pruning would say "Failed..." even if it was successful \ No newline at end of file diff --git a/changelog.txt b/changelog.txt index 16b5a7dc..fa411617 100644 --- a/changelog.txt +++ b/changelog.txt @@ -1,3 +1,21 @@ +SD GUI 1.9.1 Changelog: +- New: Resolution can now also be adjusted in InstructPix2Pix mode (resizes input image) +- New: Resulting aspect ratio is shown next to resolution option +- New: Right-click image menu in InstructPix2Pix has an option to copy a comparison image +- Improved: Model compatibility, any Safetensors file should now work fine after conversion +- Improved: Model converter no longer includes 1 GB safety checker in ONNX outputs +- Improved: Image Viewer now shows image resolution info in InstructPix2Pix mode +- Improved: InstructPix2Pix no longer downloads unnecessary files, now <2GB +- Fixed: InstructPix2Pix errors when trying to run it without internet connection +- Fixed: Certain options result in filenames being too long causing issues +- Fixed: Stable Diffusion model gets reloaded if amount of free RAM changed +- Fixed: Image export breaks if initialization image(s) import takes too long +- Fixed: "Open Output Folder" opens Documents folder +- Fixed: ONNX and InstructPix2Pix don't accept images with UTF8 characters in path +- Fixed: Running InvokeAI post-processing sometimes fails due to long paths +- Fixed: Issues with comma vs dot, depending on Windows region settings + + SD GUI 1.9.0 Changelog: - New: Added InstructPix2Pix (Use with Settings -> Implementation -> InstructPix2Pix) - New: Added the option to show the input image next to the output for comparisons diff --git a/meta/version.json b/meta/version.json index 517a7d78..2fe9e294 100644 --- a/meta/version.json +++ b/meta/version.json @@ -1,4 +1,4 @@ { - "latestPublicRel": "1.9.0", + "latestPublicRel": "1.9.1", "latestBetaRel": "1.8.1b3", } \ No newline at end of file