Type alias Options

Type declaration

• Optional CLIP_stop_at_last_layers?: number

  Clip skip

• Optional ESRGAN_tile?: number

  Tile size for ESRGAN upscalers.

• Optional ESRGAN_tile_overlap?: number

  Tile overlap for ESRGAN upscalers.

• Optional add_model_hash_to_info?: boolean

  Add model hash to infotext

• Optional add_model_name_to_info?: boolean

  Add model name to infotext

• Optional add_user_name_to_info?: any

  Add user name to infotext when authenticated

• Optional add_vae_hash_to_info?: boolean

  Add VAE hash to infotext

• Optional add_vae_name_to_info?: boolean

  Add VAE name to infotext

• Optional add_version_to_infotext?: boolean

  Add program version to infotext

• Optional always_discard_next_to_last_sigma?: any

  Always discard next-to-last sigma

• Optional api_enable_requests?: boolean

  Allow http:// and https:// URLs for input images in API

• Optional api_forbid_local_requests?: boolean

  Forbid URLs to local resources

• Optional api_useragent?: any

  User agent for requests

• Optional auto_launch_browser?: string

  Automatically open webui in browser on startup

• Optional auto_vae_precision?: boolean

  Automatically revert VAE to 32-bit floats

• Optional batch_cond_uncond?: boolean

  Batch cond/uncond

• Optional canvas_auto_expand?: boolean

  Automatically expands an image that does not fit completely in the canvas area, similar to manually pressing the S and R buttons

• Optional canvas_blur_prompt?: any

  Take the focus off the prompt when working with a canvas

• Optional canvas_disabled_functions?: any[]

  Disable function that you don't use

• Optional canvas_hotkey_adjust?: string

  Adjust brush size

• Optional canvas_hotkey_fullscreen?: string

  Fullscreen Mode, maximizes the picture so that it fits into the screen and stretches it to its full width

• Optional canvas_hotkey_move?: string

  Moving the canvas

• Optional canvas_hotkey_overlap?: string

  Toggle overlap

• Optional canvas_hotkey_reset?: string

  Reset zoom and canvas positon

• Optional canvas_hotkey_zoom?: string

  Zoom canvas

• Optional canvas_show_tooltip?: boolean

  Enable tooltip on the canvas

• Optional clean_temp_dir_at_start?: any

  Cleanup non-default temporary directory when starting webui

• Optional code_former_weight?: number

  CodeFormer weight

• Optional comma_padding_backtrack?: number

  Prompt word wrap length limit

• Optional compact_prompt_box?: any

  Compact prompt layout

• Optional cross_attention_optimization?: string

  Cross attention optimization

• Optional dataset_filename_join_string?: string

  Filename join string

• Optional dataset_filename_word_regex?: any

  Filename word regex

• Optional ddim_discretize?: string

  img2img DDIM discretize

• Optional deepbooru_escape?: boolean

  deepbooru: escape () brackets

• Optional deepbooru_filter_tags?: any

  deepbooru: filter out those tags

• Optional deepbooru_sort_alpha?: boolean

  deepbooru: sort tags alphabetically

• Optional deepbooru_use_spaces?: boolean

  deepbooru: use spaces in tags

• Optional dimensions_and_batch_together?: boolean

  Show Width/Height and Batch sliders in same row

• Optional directories_filename_pattern?: string

  Directory name pattern

• Optional directories_max_prompt_words?: number

  Max prompt words for [prompt_words] pattern

• Optional disable_all_extensions?: string

  Disable all extensions (preserves the list of disabled extensions)

• Optional disable_mmap_load_safetensors?: any

  Disable memmapping for loading .safetensors files.

• Optional disable_token_counters?: any

  Disable prompt token counters

• Optional disable_weights_auto_swap?: boolean

  Disregard checkpoint information from pasted infotext

• Optional disabled_extensions?: any

  Disable these extensions

• Optional do_not_show_images?: any

  Do not show any images in gallery

• Optional dont_fix_second_order_samplers_schedule?: any

  Do not fix prompt schedule for second order samplers.

• Optional dump_stacks_on_signal?: any

  Print stack traces before exiting the program with ctrl+c.

• Optional enable_batch_seeds?: boolean

  Make K-diffusion samplers produce same images in a batch as when making a single image

• Optional enable_console_prompts?: any

  Print prompts to console when generating with txt2img and img2img.

• Optional enable_emphasis?: boolean

  Enable emphasis

• Optional enable_pnginfo?: boolean

  Write infotext to metadata of the generated image

• Optional enable_quantization?: any

  Enable quantization in K samplers for sharper and cleaner results. This may change existing seeds

• Optional eta_ancestral?: number

  Eta for k-diffusion samplers

• Optional eta_ddim?: any

  Eta for DDIM

• Optional eta_noise_seed_delta?: any

  Eta noise seed delta

• Optional export_for_4chan?: boolean

  Save copy of large images as JPG

• Optional extra_networks_add_text_separator?: string

  Extra networks separator

• Optional extra_networks_card_height?: any

  Card height for Extra Networks

• Optional extra_networks_card_order?: string

  Default order for Extra Networks cards

• Optional extra_networks_card_order_field?: string

  Default order field for Extra Networks cards

• Optional extra_networks_card_show_desc?: boolean

  Show description on card

• Optional extra_networks_card_text_scale?: number

  Card text scale

• Optional extra_networks_card_width?: any

  Card width for Extra Networks

• Optional extra_networks_default_multiplier?: number

  Default multiplier for extra networks

• Optional extra_networks_dir_button_function?: any

  Add a '/' to the beginning of directory buttons

• Optional extra_networks_hidden_models?: string

  Show cards for models in hidden directories

• Optional extra_networks_show_hidden_directories?: boolean

  Show hidden directories

• Optional extra_options_accordion?: any

  Place added settings into an accordion

• Optional extra_options_cols?: number

  Number of columns for added settings

• Optional extra_options_img2img?: any

  Settings for img2img

• Optional extra_options_txt2img?: any

  Settings for txt2img

• Optional face_restoration?: any

  Restore faces

• Optional face_restoration_model?: string

  Face restoration model

• Optional face_restoration_unload?: any

  Move face restoration model from VRAM into RAM after processing

• Optional font?: any

  Font for image grids that have text

• Optional gallery_height?: any

  Gallery height

• Optional gradio_theme?: string

  Gradio theme

• Optional gradio_themes_cache?: boolean

  Cache gradio themes locally

• Optional grid_background_color?: string

  Background color for image grids

• Optional grid_extended_filename?: any

  Add extended info (seed, prompt) to filename when saving grid

• Optional grid_format?: string

  File format for grids

• Optional grid_only_if_multiple?: boolean

  Do not save grids consisting of one picture

• Optional grid_prevent_empty_spots?: any

  Prevent empty spots in grid (when set to autodetect)

• Optional grid_save?: boolean

  Always save all generated image grids

• Optional grid_save_to_dirs?: boolean

  Save grids to a subdirectory

• Optional grid_text_active_color?: string

  Text color for image grids

• Optional grid_text_inactive_color?: string

  Inactive text color for image grids

• Optional grid_zip_filename_pattern?: any

  Archive filename pattern

• Optional hidden_tabs?: any

  Hidden UI tabs

• Optional hide_ldm_prints?: boolean

  Prevent Stability-AI's ldm/sgm modules from printing noise to console.

• Optional hide_samplers?: any

  Hide samplers in user interface

• Optional hires_fix_refiner_pass?: string

  Hires fix: which pass to enable refiner for

• Optional hires_fix_show_prompts?: any

  Hires fix: show hires prompt and negative prompt

• Optional hires_fix_show_sampler?: any

  Hires fix: show hires checkpoint and sampler selection

• Optional hires_fix_use_firstpass_conds?: any

  For hires fix, calculate conds of second pass using extra networks of first pass.

• Optional img2img_background_color?: string

  With img2img, fill transparent parts of the input image with this color.

• Optional img2img_batch_show_results_limit?: number

  Show the first N batch img2img results in UI

• Optional img2img_color_correction?: any

  Apply color correction to img2img results to match original colors.

• Optional img2img_editor_height?: number

  Height of the image editor

• Optional img2img_extra_noise?: any

  Extra noise multiplier for img2img and hires fix

• Optional img2img_fix_steps?: any

  With img2img, do exactly the amount of steps the slider specifies.

• Optional img2img_inpaint_mask_brush_color?: string

  Inpaint mask brush color

• Optional img2img_inpaint_sketch_default_brush_color?: string

  Inpaint sketch initial brush color

• Optional img2img_settings_accordion?: any

  Settings in img2img hidden under Accordion

• Optional img2img_sketch_default_brush_color?: string

  Sketch initial brush color

• Optional img_downscale_threshold?: number

  File size limit for the above option, MB

• Optional img_max_size_mp?: number

  Maximum image size

• Optional infotext_explanation?: string

• Optional infotext_skip_pasting?: any

  Disregard fields from pasted infotext

• Optional infotext_styles?: string

  Infer styles from prompts of pasted infotext

• Optional initial_noise_multiplier?: number

  Noise multiplier for img2img

• Optional inpainting_mask_weight?: number

  Inpainting conditioning mask strength

• Optional interrogate_clip_dict_limit?: number

  CLIP: maximum number of lines in text file

• Optional interrogate_clip_max_length?: number

  BLIP: maximum description length

• Optional interrogate_clip_min_length?: number

  BLIP: minimum description length

• Optional interrogate_clip_num_beams?: number

  BLIP: num_beams

• Optional interrogate_clip_skip_categories?: any

  CLIP: skip inquire categories

• Optional interrogate_deepbooru_score_threshold?: number

  deepbooru: score threshold

• Optional interrogate_keep_models_in_memory?: any

  Keep models in VRAM

• Optional interrogate_return_ranks?: any

  Include ranks of model tags matches in results.

• Optional jpeg_quality?: number

  Quality for saved jpeg images

• Optional js_live_preview_in_modal_lightbox?: any

  Show Live preview in full page image viewer

• Optional js_modal_lightbox?: boolean

  Full page image viewer: enable

• Optional js_modal_lightbox_gamepad?: any

  Full page image viewer: navigate with gamepad

• Optional js_modal_lightbox_gamepad_repeat?: number

  Full page image viewer: gamepad repeat period

• Optional js_modal_lightbox_initially_zoomed?: boolean

  Full page image viewer: show images zoomed in by default

• Optional k_sched_type?: string

  Scheduler type

• Optional keyedit_delimiters?: string

  Word delimiters when editing the prompt with Ctrl+up/down

• Optional keyedit_delimiters_whitespace?: any[]

  Ctrl+up/down whitespace delimiters

• Optional keyedit_move?: boolean

  Alt+left/right moves prompt elements

• Optional keyedit_precision_attention?: number

  Precision for (attention:1.1) when editing the prompt with Ctrl+up/down

• Optional keyedit_precision_extra?: number

  Precision for when editing the prompt with Ctrl+up/down

• Optional list_hidden_files?: boolean

  Load models/files in hidden directories

• Optional live_preview_allow_lowvram_full?: any

  Allow Full live preview method with lowvram/medvram

• Optional live_preview_content?: string

  Live preview subject

• Optional live_preview_fast_interrupt?: any

  Return image with chosen live preview method on interrupt

• Optional live_preview_refresh_period?: number

  Progressbar and preview update period

• Optional live_previews_enable?: boolean

  Show live previews of the created image

• Optional live_previews_image_format?: string

  Live preview file format

• Optional localization?: string

  Localization

• Optional lora_add_hashes_to_infotext?: boolean

  Add Lora hashes to infotext

• Optional lora_functional?: any

  Lora/Networks: use old method that takes longer when you have multiple Loras active and produces same results as kohya-ss/sd-webui-additional-networks extension

• Optional lora_hide_unknown_for_versions?: any

  Hide networks of unknown versions for model versions

• Optional lora_in_memory_limit?: any

  Number of Lora networks to keep cached in memory

• Optional lora_preferred_name?: string

  When adding to prompt, refer to Lora by

• Optional lora_show_all?: any

  Always show all networks on the Lora page

• Optional memmon_poll_rate?: number

  VRAM usage polls per second during generation.

• Optional multiple_tqdm?: boolean

  Add a second progress bar to the console that shows progress for an entire job.

• Optional n_rows?: number

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

• Optional no_dpmpp_sde_batch_determinism?: any

  Do not make DPM++ SDE deterministic across different batch sizes.

• Optional notification_audio?: boolean

  Play notification sound after image generation

• Optional notification_volume?: number

  Notification sound volume

• Optional outdir_extras_samples?: string

  Output directory for images from extras tab

• Optional outdir_grids?: any

  Output directory for grids; if empty, defaults to two directories below

• Optional outdir_img2img_grids?: string

  Output directory for img2img grids

• Optional outdir_img2img_samples?: string

  Output directory for img2img images

• Optional outdir_init_images?: string

  Directory for saving init images when using img2img

• Optional outdir_samples?: any

  Output directory for images; if empty, defaults to three directories below

• Optional outdir_save?: string

  Directory for saving images using the Save button

• Optional outdir_txt2img_grids?: string

  Output directory for txt2img grids

• Optional outdir_txt2img_samples?: string

  Output directory for txt2img images

• Optional pad_cond_uncond?: any

  Pad prompt/negative prompt to be same length

• Optional persistent_cond_cache?: boolean

  Persistent cond cache

• Optional pin_memory?: any

  Turn on pin_memory for DataLoader. Makes training slightly faster but can increase memory usage.

• Optional postprocessing_enable_in_main_ui?: any

  Enable postprocessing operations in txt2img and img2img tabs

• Optional postprocessing_existing_caption_action?: string

  Action for existing captions

• Optional postprocessing_operation_order?: any

  Postprocessing operation order

• Optional print_hypernet_extra?: any

  Print extra hypernetwork information to console.

• Optional quicksettings_list?: any[]

  Quicksettings list

• Optional randn_source?: string

  Random number generator source.

• Optional realesrgan_enabled_models?: any[]

  Select which Real-ESRGAN models to show in the web UI.

• Optional restore_config_state_file?: any

  Config state file to restore from, under 'config-states/' folder

• Optional return_grid?: boolean

  Show grid in gallery

• Optional return_mask?: any

  For inpainting, include the greyscale mask in results for web

• Optional return_mask_composite?: any

  For inpainting, include masked composite in results for web

• Optional rho?: any

  rho

• Optional s_churn?: any

  sigma churn

• Optional s_min_uncond?: any

  Negative Guidance minimum sigma

• Optional s_noise?: number

  sigma noise

• Optional s_tmax?: any

  sigma tmax

• Optional s_tmin?: any

  sigma tmin

• Optional samplers_in_dropdown?: boolean

  Use dropdown for sampler selection instead of radio group

• Optional samples_filename_pattern?: any

  Images filename pattern

• Optional samples_format?: string

  File format for images

• Optional samples_log_stdout?: any

  Always print all generation info to standard output

• Optional samples_save?: boolean

  Always save all generated images

• Optional save_images_add_number?: boolean

  Add number to filename when saving

• Optional save_images_before_color_correction?: any

  Save a copy of image before applying color correction to img2img results

• Optional save_images_before_face_restoration?: any

  Save a copy of image before doing face restoration.

• Optional save_images_before_highres_fix?: any

  Save a copy of image before applying highres fix.

• Optional save_images_replace_action?: string

  Saving the image to an existing file

• Optional save_incomplete_images?: any

  Save incomplete images

• Optional save_init_img?: any

  Save init images when using img2img

• Optional save_mask?: any

  For inpainting, save a copy of the greyscale mask

• Optional save_mask_composite?: any

  For inpainting, save a masked composite

• Optional save_optimizer_state?: any

  Saves Optimizer state as separate *.optim file. Training of embedding or HN can be resumed with the matching optim file.

• Optional save_selected_only?: boolean

  When using 'Save' button, only save a single selected image

• Optional save_to_dirs?: boolean

  Save images to a subdirectory

• Optional save_training_settings_to_txt?: boolean

  Save textual inversion and hypernet settings to a text file whenever training starts.

• Optional save_txt?: any

  Create a text file with infotext next to every generated image

• Optional sd_checkpoint_cache?: any

  Checkpoints to cache in RAM

• Optional sd_checkpoint_dropdown_use_short?: any

  Checkpoint dropdown: use filenames without paths

• Optional sd_checkpoint_hash?: any

  SHA256 hash of the current checkpoint

• Optional sd_checkpoints_keep_in_cpu?: boolean

  Only keep one model on device

• Optional sd_checkpoints_limit?: number

  Maximum number of checkpoints loaded at the same time

• Optional sd_hypernetwork?: string

  Add hypernetwork to prompt

• Optional sd_lora?: string

  Add network to prompt

• Optional sd_model_checkpoint?: any

  Stable Diffusion checkpoint

• Optional sd_unet?: string

  SD Unet

• Optional sd_vae?: string

  SD VAE

• Optional sd_vae_checkpoint_cache?: any

  VAE Checkpoints to cache in RAM

• Optional sd_vae_decode_method?: string

  VAE type for decode

• Optional sd_vae_encode_method?: string

  VAE type for encode

• Optional sd_vae_explanation?: string

• Optional sd_vae_overrides_per_model_preferences?: boolean

  Selected VAE overrides per-model preferences

• Optional sdxl_crop_left?: any

  crop left coordinate

• Optional sdxl_crop_top?: any

  crop top coordinate

• Optional sdxl_refiner_high_aesthetic_score?: number

  SDXL high aesthetic score

• Optional sdxl_refiner_low_aesthetic_score?: number

  SDXL low aesthetic score

• Optional send_seed?: boolean

  Send seed when sending prompt or image to other interface

• Optional send_size?: boolean

  Send size when sending prompt or image to another interface

• Optional settings_in_ui?: string

• Optional sgm_noise_multiplier?: any

  SGM noise multiplier

• Optional show_gradio_deprecation_warnings?: boolean

  Show gradio deprecation warnings in console.

• Optional show_progress_every_n_steps?: number

  Live preview display period

• Optional show_progress_grid?: boolean

  Show previews of all images generated in a batch as a grid

• Optional show_progress_in_title?: boolean

  Show generation progress in window title.

• Optional show_progress_type?: string

  Live preview method

• Optional show_progressbar?: boolean

  Show progressbar

• Optional show_warnings?: any

  Show warnings in console.

• Optional sigma_max?: any

  sigma max

• Optional sigma_min?: any

  sigma min

• Optional target_side_length?: number

  Width/height limit for the above option, in pixels

• Optional temp_dir?: any

  Directory for temporary images; leave empty for default

• Optional textual_inversion_add_hashes_to_infotext?: boolean

  Add Textual Inversion hashes to infotext

• Optional textual_inversion_print_at_load?: any

  Print a list of Textual Inversion embeddings when loading model

• Optional tiling?: any

  Tiling

• Optional token_merging_ratio?: any

  Token merging ratio

• Optional token_merging_ratio_hr?: any

  Token merging ratio for high-res pass

• Optional token_merging_ratio_img2img?: any

  Token merging ratio for img2img

• Optional training_enable_tensorboard?: any

  Enable tensorboard logging.

• Optional training_image_repeats_per_epoch?: number

  Number of repeats for a single input image per epoch; used only for displaying epoch number

• Optional training_tensorboard_flush_every?: number

  How often, in seconds, to flush the pending tensorboard events and summaries to disk.

• Optional training_tensorboard_save_images?: any

  Save generated images within tensorboard.

• Optional training_write_csv_every?: number

  Save an csv containing the loss to log directory every N steps, 0 to disable

• Optional training_xattention_optimizations?: any

  Use cross attention optimizations while training

• Optional txt2img_settings_accordion?: any

  Settings in txt2img hidden under Accordion

• Optional ui_extra_networks_tab_reorder?: any

  Extra networks tab order

• Optional ui_reorder_list?: any

  UI item order for txt2img/img2img tabs

• Optional ui_tab_order?: any

  UI tab order

• Optional uni_pc_lower_order_final?: boolean

  UniPC lower order final

• Optional uni_pc_order?: number

  UniPC order

• Optional uni_pc_skip_type?: string

  UniPC skip type

• Optional uni_pc_variant?: string

  UniPC variant

• Optional unload_models_when_training?: any

  Move VAE and CLIP to RAM when training if possible. Saves VRAM.

• Optional upcast_attn?: any

  Upcast cross attention layer to float32

• Optional upscaler_for_img2img?: any

  Upscaler for img2img

• Optional upscaling_max_images_in_cache?: number

  Maximum number of images in upscaling cache

• Optional use_old_emphasis_implementation?: any

  Use old emphasis implementation. Can be useful to reproduce old seeds.

• Optional use_old_hires_fix_width_height?: any

  For hires fix, use width/height sliders to set final resolution rather than first pass (disables Upscale by, Resize width/height to).

• Optional use_old_karras_scheduler_sigmas?: any

  Use old karras scheduler sigmas (0.1 to 10).

• Optional use_old_scheduling?: any

  Use old prompt editing timelines.

• Optional use_original_name_batch?: boolean

  Use original name for output filename during batch process in extras tab

• Optional use_save_to_dirs_for_ui?: any

  When using "Save" button, save images to a subdirectory

• Optional use_upscaler_name_as_suffix?: any

  Use upscaler name as filename suffix in the extras tab

• Optional webp_lossless?: any

  Use lossless compression for webp images