Samples Save Always save all generated images default Value true

Samples Format File format for images default Value "png"

Samples Filename Pattern Images filename pattern default Value ""

Save Images Add Number Add number to filename when saving default Value true

Grid Save Always save all generated image grids default Value true

Grid Format File format for grids default Value "png"

Grid Extended Filename Add extended info (seed, prompt) to filename when saving grid default Value false

Grid Only If Multiple Do not save grids consisting of one picture default Value true

Grid Prevent Empty Spots Prevent empty spots in grid (when set to autodetect) default Value false

N Rows Grid row count; use -1 for autodetect and 0 for it to be same as batch size default Value -1

Enable Pnginfo Save text information about generation parameters as chunks to png files default Value true

Save Txt Create a text file next to every image with generation parameters. default Value false

Save Images Before Face Restoration Save a copy of image before doing face restoration. default Value false

Save Images Before Highres Fix Save a copy of image before applying highres fix. default Value false

Save Images Before Color Correction Save a copy of image before applying color correction to img2img results default Value false

Jpeg Quality Quality for saved jpeg images default Value 80

Export For 4Chan If PNG image is larger than 4MB or any dimension is larger than 4000, downscale and save copy as JPG default Value true

Use Original Name Batch Use original name for output filename during batch process in extras tab default Value false

Save Selected Only When using 'Save' button, only save a single selected image default Value true

Do Not Add Watermark Do not add watermark to images default Value false

Temp Dir Directory for temporary images; leave empty for default default Value ""

Clean Temp Dir At Start Cleanup non-default temporary directory when starting webui default Value false

Outdir Samples Output directory for images; if empty, defaults to three directories below default Value ""

Outdir Txt2Img Samples Output directory for txt2img images default Value "outputs/txt2img-images"

Outdir Img2Img Samples Output directory for img2img images default Value "outputs/img2img-images"

Outdir Extras Samples Output directory for images from extras tab default Value "outputs/extras-images"

Outdir Grids Output directory for grids; if empty, defaults to two directories below default Value ""

Outdir Txt2Img Grids Output directory for txt2img grids default Value "outputs/txt2img-grids"

Outdir Img2Img Grids Output directory for img2img grids default Value "outputs/img2img-grids"

Outdir Save Directory for saving images using the Save button default Value "log/images"

Save To Dirs Save images to a subdirectory default Value false

Grid Save To Dirs Save grids to a subdirectory default Value false

Use Save To Dirs For Ui When using "Save" button, save images to a subdirectory default Value false

Directories Filename Pattern Directory name pattern default Value ""

Directories Max Prompt Words Max prompt words for [prompt_words] pattern default Value 8

Esrgan Tile Tile size for ESRGAN upscalers. 0 = no tiling. default Value 192

Esrgan Tile Overlap Tile overlap, in pixels for ESRGAN upscalers. Low values = visible seam. default Value 8

Realesrgan Enabled Models Select which Real-ESRGAN models to show in the web UI. (Requires restart) default Value ["R-ESRGAN 4x+","R-ESRGAN 4x+ Anime6B"]

Upscaler For Img2Img Upscaler for img2img

Use Scale Latent For Hires Fix Upscale latent space image when doing hires. fix default Value false

Ldsr Steps LDSR processing steps. Lower = faster default Value 100

Swin Tile Tile size for all SwinIR. default Value 192

Swin Tile Overlap Tile overlap, in pixels for SwinIR. Low values = visible seam. default Value 8

Face Restoration Model Face restoration model

Code Former Weight CodeFormer weight parameter; 0 = maximum effect; 1 = minimum effect default Value 0.5

Face Restoration Unload Move face restoration model from VRAM into RAM after processing default Value false

Memmon Poll Rate VRAM usage polls per second during generation. Set to 0 to disable. default Value 8

Samples Log Stdout Always print all generation info to standard output default Value false

Multiple Tqdm Add a second progress bar to the console that shows progress for an entire job. default Value true

Unload Models When Training Move VAE and CLIP to RAM when training if possible. Saves VRAM. default Value false

Pin Memory Turn on pin_memory for DataLoader. Makes training slightly faster but can increase memory usage. default Value false

Save Optimizer State Saves Optimizer state as separate *.optim file. Training can be resumed with HN itself and matching optim file. default Value false

Dataset Filename Word Regex Filename word regex default Value ""

Dataset Filename Join String Filename join string default Value " "

Training Image Repeats Per Epoch Number of repeats for a single input image per epoch; used only for displaying epoch number default Value 1

Training Write Csv Every Save an csv containing the loss to log directory every N steps, 0 to disable default Value 500

Training Xattention Optimizations Use cross attention optimizations while training default Value false

Sd Model Checkpoint Stable Diffusion checkpoint

Sd Checkpoint Cache Checkpoints to cache in RAM default Value 0

Sd Vae SD VAE default Value "auto"

Sd Vae As Default Ignore selected VAE for stable diffusion checkpoints that have their own .vae.pt next to them default Value false

Sd Hypernetwork Hypernetwork default Value "None"

Sd Hypernetwork Strength Hypernetwork strength default Value 1

Inpainting Mask Weight Inpainting conditioning mask strength default Value 1

Img2Img Color Correction Apply color correction to img2img results to match original colors. default Value false

Img2Img Fix Steps With img2img, do exactly the amount of steps the slider specifies (normally you'd do less with less denoising). default Value false

Enable Quantization Enable quantization in K samplers for sharper and cleaner results. This may change existing seeds. Requires restart to apply. default Value false

Enable Emphasis Emphasis: use (text) to make model pay more attention to text and [text] to make it pay less attention default Value true

Use Old Emphasis Implementation Use old emphasis implementation. Can be useful to reproduce old seeds. default Value false

Enable Batch Seeds Make K-diffusion samplers produce same images in a batch as when making a single image default Value true

Comma Padding Backtrack Increase coherency by padding from the last comma within n tokens when using more than 75 tokens default Value 20

Filter Nsfw Filter NSFW content default Value false

Clip Stop At Last Layers Clip skip default Value 1

Random Artist Categories Allowed categories for random artists selection when using the Roll button default Value []

Interrogate Keep Models In Memory Interrogate: keep models in VRAM default Value false

Interrogate Use Builtin Artists Interrogate: use artists from artists.csv default Value true

Interrogate Return Ranks Interrogate: include ranks of model tags matches in results (Has no effect on caption-based interrogators). default Value false

Interrogate Clip Num Beams Interrogate: num_beams for BLIP default Value 1

Interrogate Clip Min Length Interrogate: minimum description length (excluding artists, etc..) default Value 24

Interrogate Clip Max Length Interrogate: maximum description length default Value 48

Interrogate Clip Dict Limit CLIP: maximum number of lines in text file (0 = No limit) default Value 1500

Interrogate Deepbooru Score Threshold Interrogate: deepbooru score threshold default Value 0.5

Deepbooru Sort Alpha Interrogate: deepbooru sort alphabetically default Value true

Deepbooru Use Spaces use spaces for tags in deepbooru default Value false

Deepbooru Escape escape () brackets in deepbooru (so they are used as literal brackets and not for emphasis) default Value true

Show Progressbar Show progressbar default Value true

Show Progress Every N Steps Show image creation progress every N sampling steps. Set to 0 to disable. Set to -1 to show after completion of batch. default Value 0

Show Progress Grid Show previews of all images generated in a batch as a grid default Value true

Return Grid Show grid in results for web default Value true

Do Not Show Images Do not show any images in results for web default Value false

Add Model Hash To Info Add model hash to generation information default Value true

Add Model Name To Info Add model name to generation information default Value false

Disable Weights Auto Swap When reading generation parameters from text into UI (from PNG info or pasted text), do not change the selected model/checkpoint. default Value false

Send Seed Send seed when sending prompt or image to other interface default Value true

Font Font for image grids that have text default Value ""

Js Modal Lightbox Enable full page image viewer default Value true

Js Modal Lightbox Initially Zoomed Show images zoomed in by default in full page image viewer default Value true

Show Progress In Title Show generation progress in window title. default Value true

Quicksettings Quicksettings list default Value "sd_model_checkpoint"

Localization Localization (requires restart) default Value "None"

Hide Samplers Hide samplers in user interface (requires restart) default Value []

Eta Ddim eta (noise multiplier) for DDIM default Value 0

Eta Ancestral eta (noise multiplier) for ancestral samplers default Value 1

Ddim Discretize img2img DDIM discretize default Value "uniform"

S Churn sigma churn default Value 0

S Tmin sigma tmin default Value 0

S Noise sigma noise default Value 1

Eta Noise Seed Delta Eta noise seed delta default Value 0

Disabled Extensions Disable those extensions default Value []