option to show batch img2img results in UI (#14009)
infotext updates: add option to disregard certain infotext fields, add option to not include VAE in infotext, add explanation to infotext settings page, move some options to infotext settings page
add FP32 fallback support on sd_vae_approx (#14046)
support XYZ scripts / split hires path from unet (#14126)
make extra network card description plaintext by default, with an option (Treat card description as HTML) to re-enable HTML as it was (originally by #13241)
rework DDIM, PLMS, UniPC to use CFG denoiser same as in k-diffusion samplers:
makes all of them work with img2img
makes prompt composition possible (AND)
makes them available for SDXL
always show extra networks tabs in the UI (#11808)
use less RAM when creating models (#11958, #12599)
textual inversion inference support for SDXL
extra networks UI: show metadata for SD checkpoints
checkpoint merger: add metadata support
prompt editing and attention: add support for whitespace after the number ([ red : green : 0.5 ]) (seed breaking change) (#12177)
VAE: allow selecting own VAE for each checkpoint (in user metadata editor)
VAE: add selected VAE to infotext
options in main UI: add own separate setting for txt2img and img2img, correctly read values from pasted infotext, add setting for column count (#12551)
add resize handle to txt2img and img2img tabs, allowing to change the amount of horizontable space given to generation parameters and resulting image gallery (#12687, #12723)
change default behavior for batching cond/uncond -- now it's on by default, and is disabled by an UI setting (Optimizatios -> Batch cond/uncond) - if you are on lowvram/medvram and are getting OOM exceptions, you will need to enable it
show current position in queue and make it so that requests are processed in the order of arrival (#12707)
add --medvram-sdxl flag that only enables --medvram for SDXL models
prompt editing timeline has separate range for first pass and hires-fix pass (seed breaking change) (#12457)
don't show hidden samplers in dropdown for XYZ script (#12780)
fix style editing dialog breaking if it's opened in both img2img and txt2img tabs
fix a bug allowing users to bypass gradio and API authentication (reported by vysecurity)
fix notification not playing when built-in webui tab is inactive (#12834)
honor --skip-install for extension installers (#12832)
don't print blank stdout in extension installers (#12833, #12855)
do not change quicksettings dropdown option when value returned is None (#12854)
get progressbar to display correctly in extensions tab
1.5.2
Bug Fixes:
fix memory leak when generation fails
update doggettx cross attention optimization to not use an unreasonable amount of memory in some edge cases -- suggestion by MorkTheOrk
1.5.1
Minor:
support parsing text encoder blocks in some new LoRAs
delete scale checker script due to user demand
Extensions and API:
add postprocess_batch_list script callback
Bug Fixes:
fix TI training for SD1
fix reload altclip model error
prepend the pythonpath instead of overriding it
fix typo in SD_WEBUI_RESTARTING
if txt2img/img2img raises an exception, finally call state.end()
fix composable diffusion weight parsing
restyle Startup profile for black users
fix webui not launching with --nowebui
catch exception for non git extensions
fix some options missing from /sdapi/v1/options
fix for extension update status always saying "unknown"
fix display of extra network cards that have <> in the name
update lora extension to work with python 3.8
1.5.0
Features:
SD XL support
user metadata system for custom networks
extended Lora metadata editor: set activation text, default weight, view tags, training info
Lora extension rework to include other types of networks (all that were previously handled by LyCORIS extension)
show github stars for extensions
img2img batch mode can read extra stuff from png info
img2img batch works with subdirectories
hotkeys to move prompt elements: alt+left/right
restyle time taken/VRAM display
add textual inversion hashes to infotext
optimization: cache git extension repo information
move generate button next to the generated picture for mobile clients
hide cards for networks of incompatible Stable Diffusion version in Lora extra networks interface
skip installing packages with pip if they all are already installed - startup speedup of about 2 seconds
Minor:
checkbox to check/uncheck all extensions in the Installed tab
add gradio user to infotext and to filename patterns
allow gif for extra network previews
add options to change colors in grid
use natural sort for items in extra networks
Mac: use empty_cache() from torch 2 to clear VRAM
added automatic support for installing the right libraries for Navi3 (AMD)
add option SWIN_torch_compile to accelerate SwinIR upscale
suppress printing TI embedding info at start to console by default
speedup extra networks listing
added [none] filename token.
removed thumbs extra networks view mode (use settings tab to change width/height/scale to get thumbs)
add always_discard_next_to_last_sigma option to XYZ plot
automatically switch to 32-bit float VAE if the generated picture has NaNs without the need for --no-half-vae commandline flag.
Extensions and API:
api endpoints: /sdapi/v1/server-kill, /sdapi/v1/server-restart, /sdapi/v1/server-stop
allow Script to have custom metaclass
add model exists status check /sdapi/v1/options
rename --add-stop-route to --api-server-stop
add before_hr script callback
add callback after_extra_networks_activate
disable rich exception output in console for API by default, use WEBUI_RICH_EXCEPTIONS env var to enable
return http 404 when thumb file not found
allow replacing extensions index with environment variable
Bug Fixes:
fix for catch errors when retrieving extension index #11290
fix very slow loading speed of .safetensors files when reading from network drives
API cache cleanup
fix UnicodeEncodeError when writing to file CLIP Interrogator batch mode
fix warning of 'has_mps' deprecated from PyTorch
fix problem with extra network saving images as previews losing generation info
fix throwing exception when trying to resize image with I;16 mode
fix for #11534: canvas zoom and pan extension hijacking shortcut keys
fixed launch script to be runnable from any directory
don't add "Seed Resize: -1x-1" to API image metadata
correctly remove end parenthesis with ctrl+up/down
fixing --subpath on newer gradio version
fix: check fill size none zero when resize (fixes #11425)
use submit and blur for quick settings textbox
save img2img batch with images.save_image()
prevent running preload.py for disabled extensions
fix: previously, model name was added together with directory name to infotext and to [model_name] filename pattern; directory name is now not included
1.4.1
Bug Fixes:
add queue lock for refresh-checkpoints
1.4.0
Features:
zoom controls for inpainting
run basic torch calculation at startup in parallel to reduce the performance impact of first generation
option to pad prompt/neg prompt to be same length
remove taming_transformers dependency
custom k-diffusion scheduler settings
add an option to show selected settings in main txt2img/img2img UI
sysinfo tab in settings
infer styles from prompts when pasting params into the UI
an option to control the behavior of the above
Minor:
bump Gradio to 3.32.0
bump xformers to 0.0.20
Add option to disable token counters
tooltip fixes & optimizations
make it possible to configure filename for the zip download
[vae_filename] pattern for filenames
Revert discarding penultimate sigma for DPM-Solver++(2M) SDE
change UI reorder setting to multiselect
read version info form CHANGELOG.md if git version info is not available
link footer API to Wiki when API is not active
persistent conds cache (opt-in optimization)
Extensions:
After installing extensions, webui properly restarts the process rather than reloads the UI
Added VAE listing to web API. Via: /sdapi/v1/sd-vae
custom unet support
Add onAfterUiUpdate callback
refactor EmbeddingDatabase.register_embedding() to allow unregistering
add before_process callback for scripts
add ability for alwayson scripts to specify section and let user reorder those sections
Bug Fixes:
Fix dragging text to prompt
fix incorrect quoting for infotext values with colon in them
fix "hires. fix" prompt sharing same labels with txt2img_prompt
Fix s_min_uncond default type int
Fix for #10643 (Inpainting mask sometimes not working)
fix bad styling for thumbs view in extra networks #10639
small fixes to prepare_tcmalloc for Debian/Ubuntu compatibility
fix --ui-debug-mode exit
patch GitPython to not use leaky persistent processes
fix duplicate Cross attention optimization after UI reload
torch.cuda.is_available() check for SdOptimizationXformers
fix hires fix using wrong conds in second pass if using Loras.
handle exception when parsing generation parameters from png info
fix upcast attention dtype error
forcing Torch Version to 1.13.1 for RX 5000 series GPUs
split mask blur into X and Y components, patch Outpainting MK2 accordingly
don't die when a LoRA is a broken symlink
allow activation of Generate Forever during generation
1.3.2
Bug Fixes:
fix files served out of tmp directory even if they are saved to disk
fix postprocessing overwriting parameters
1.3.1
Features:
revert default cross attention optimization to Doggettx
Bug Fixes:
fix bug: LoRA don't apply on dropdown list sd_lora
fix png info always added even if setting is not enabled
fix some fields not applying in xyz plot
fix "hires. fix" prompt sharing same labels with txt2img_prompt
fix lora hashes not being added properly to infotex if there is only one lora
fix --use-cpu failing to work properly at startup
make --disable-opt-split-attention command line option work again
1.3.0
Features:
add UI to edit defaults
token merging (via dbolya/tomesd)
settings tab rework: add a lot of additional explanations and links
load extensions' Git metadata in parallel to loading the main program to save a ton of time during startup
update extensions table: show branch, show date in separate column, and show version from tags if available
TAESD - another option for cheap live previews
allow choosing sampler and prompts for second pass of hires fix - hidden by default, enabled in settings
calculate hashes for Lora
add lora hashes to infotext
when pasting infotext, use infotext's lora hashes to find local loras for <lora:xxx:1> entries whose hashes match loras the user has
select cross attention optimization from UI
Minor:
bump Gradio to 3.31.0
bump PyTorch to 2.0.1 for macOS and Linux AMD
allow setting defaults for elements in extensions' tabs
allow selecting file type for live previews
show "Loading..." for extra networks when displaying for the first time
suppress ENSD infotext for samplers that don't use it
clientside optimizations
add options to show/hide hidden files and dirs in extra networks, and to not list models/files in hidden directories
allow whitespace in styles.csv
add option to reorder tabs
move some functionality (swap resolution and set seed to -1) to client
option to specify editor height for img2img
button to copy image resolution into img2img width/height sliders
switch from pyngrok to ngrok-py
lazy-load images in extra networks UI
set "Navigate image viewer with gamepad" option to false by default, by request
change upscalers to download models into user-specified directory (from commandline args) rather than the default models/<...>
allow hiding buttons in ui-config.json
Extensions:
add /sdapi/v1/script-info api
use Ruff to lint Python code
use ESlint to lint Javascript code
add/modify CFG callbacks for Self-Attention Guidance extension
add command and endpoint for graceful server stopping
add some locals (prompts/seeds/etc) from processing function into the Processing class as fields
rework quoting for infotext items that have commas in them to use JSON (should be backwards compatible except for cases where it didn't work previously)
add /sdapi/v1/refresh-loras api checkpoint post request
tests overhaul
Bug Fixes:
fix an issue preventing the program from starting if the user specifies a bad Gradio theme
fix broken prompts from file script
fix symlink scanning for extra networks
fix --data-dir ignored when launching via webui-user.bat COMMANDLINE_ARGS
allow web UI to be ran fully offline
fix inability to run with --freeze-settings
fix inability to merge checkpoint without adding metadata
fix extra networks' save preview image not adding infotext for jpeg/webm
remove blinking effect from text in hires fix and scale resolution preview
make links to http://<...>.git extensions work in the extension tab
fix bug with webui hanging at startup due to hanging git process
1.2.1
Features:
add an option to always refer to LoRA by filenames
Bug Fixes:
never refer to LoRA by an alias if multiple LoRAs have same alias or the alias is called none
fix upscalers disappearing after the user reloads UI
allow bf16 in safe unpickler (resolves problems with loading some LoRAs)
allow web UI to be ran fully offline
fix localizations not working
fix error for LoRAs: 'LatentDiffusion' object has no attribute 'lora_layer_mapping'
1.2.0
Features:
do not wait for Stable Diffusion model to load at startup
add filename patterns: [denoising]
directory hiding for extra networks: dirs starting with . will hide their cards on extra network tabs unless specifically searched for
LoRA: for the <...> text in prompt, use name of LoRA that is in the metadata of the file, if present, instead of filename (both can be used to activate LoRA)
LoRA: read infotext params from kohya-ss's extension parameters if they are present and if his extension is not active
LoRA: fix some LoRAs not working (ones that have 3x3 convolution layer)
LoRA: add an option to use old method of applying LoRAs (producing same results as with kohya-ss)
add version to infotext, footer and console output when starting
add links to wiki for filename pattern settings
add extended info for quicksettings setting and use multiselect input instead of a text field
Minor:
bump Gradio to 3.29.0
bump PyTorch to 2.0.1
--subpath option for gradio for use with reverse proxy
Linux/macOS: use existing virtualenv if already active (the VIRTUAL_ENV environment variable)
do not apply localizations if there are none (possible frontend optimization)
add extra None option for VAE in XYZ plot
print error to console when batch processing in img2img fails
create HTML for extra network pages only on demand
allow directories starting with . to still list their models for LoRA, checkpoints, etc
put infotext options into their own category in settings tab
do not show licenses page when user selects Show all pages in settings
Extensions:
tooltip localization support
add API method to get LoRA models with prompt
Bug Fixes:
re-add /docs endpoint
fix gamepad navigation
make the lightbox fullscreen image function properly
fix squished thumbnails in extras tab
keep "search" filter for extra networks when user refreshes the tab (previously it showed everything after you refreshed)
fix webui showing the same image if you configure the generation to always save results into same file
fix bug with upscalers not working properly
fix MPS on PyTorch 2.0.1, Intel Macs
make it so that custom context menu from contextMenu.js only disappears after user's click, ignoring non-user click events
prevent Reload UI button/link from reloading the page when it's not yet ready
fix prompts from file script failing to read contents from a drag/drop file
1.1.1
Bug Fixes:
fix an error that prevents running webui on PyTorch<2.0 without --disable-safe-unpickle
add support for saving init images in img2img, and record their hashes in infotext for reproducibility
automatically select current word when adjusting weight with ctrl+up/down
add dropdowns for X/Y/Z plot
add setting: Stable Diffusion/Random number generator source: makes it possible to make images generated from a given manual seed consistent across different GPUs
support Gradio's theme API
use TCMalloc on Linux by default; possible fix for memory leaks
add optimization option to remove negative conditioning at low sigma values #9177
