Type alias Flags

Type declaration

• Optional ad_no_huggingface?: boolean

  Don't use adetailer models from huggingface

• Optional add_stop_route?: boolean

  does not do anything

• Optional administrator?: boolean

  Administrator rights

• Optional agent_scheduler_sqlite_file?: string

  sqlite file to use for the database connection. It can be abs or relative path(from base path) default: task_scheduler.sqlite3

• Optional allow_code?: boolean

  allow custom script execution from webui

• Optional always_batch_cond_uncond?: boolean

  does not do anything

• Optional api?: boolean

  use api=True to launch the API together with the webui (use --nowebui instead for only the API)

• Optional api_auth?: string

  Set authentication for API like "username:password"; or comma-delimit multiple like "u1:p1,u2:p2,u3:p3"

• Optional api_log?: boolean

  use api-log=True to enable logging of all API requests

• Optional api_server_stop?: boolean

  enable server stop/restart/kill via api

• Optional autolaunch?: boolean

  open the webui URL in the system's default browser upon launch

• Optional bsrgan_models_path?: string

  Path to directory with BSRGAN model file(s).

• Optional ckpt?: string

  path to checkpoint of stable diffusion model; if specified, this checkpoint will be added to the list of checkpoints and loaded

• Optional ckpt_dir?: string

  Path to directory with stable diffusion checkpoints

• Optional clip_models_path?: string

  Path to directory with CLIP model file(s).

• Optional codeformer_models_path?: string

  Path to directory with codeformer model file(s).

• Optional config?: string

  path to config which constructs model

• Optional controlnet_annotator_models_path?: string

  Path to directory with annotator model directories

• Optional controlnet_dir?: string

  Path to directory with ControlNet models

• Optional controlnet_loglevel?: string

  Set the log level (DEBUG, INFO, WARNING, ERROR, CRITICAL)

• Optional controlnet_preprocessor_cache_size?: number

  Cache size for controlnet preprocessor results

• Optional controlnet_tracemalloc?: string

  Enable memory tracing.

• Optional cors_allow_origins?: string

  Allowed CORS origin(s) in the form of a comma-separated list (no spaces)

• Optional cors_allow_origins_regex?: string

  Allowed CORS origin(s) in the form of a single regular expression

• Optional data_dir?: string

  base path where all user data is stored

• Optional deepdanbooru?: boolean

  does not do anything

• Optional device_id?: string

  Select the default CUDA device to use (export CUDA_VISIBLE_DEVICES=0,1,etc might be needed before)

• Optional disable_all_extensions?: boolean

  prevent all extensions from running regardless of any other settings

• Optional disable_console_progressbars?: boolean

  do not output progressbars to console

• Optional disable_extra_extensions?: boolean

  prevent all extensions except built-in from running regardless of any other settings

• Optional disable_model_loading_ram_optimization?: boolean

  disable an optimization that reduces RAM use when loading a model

• Optional disable_nan_check?: boolean

  do not check if produced images/latent spaces have nans; useful for running without a checkpoint in CI

• Optional disable_openpose_editor_auto_update?: string

  Disable auto-update of openpose editor

• Optional disable_opt_split_attention?: boolean

  prefer no cross-attention layer optimization for automatic choice of optimization

• Optional disable_safe_unpickle?: boolean

  disable checking pytorch models for malicious code

• Optional disable_tls_verify?: string

  When passed, enables the use of self-signed certificates.

• Optional do_not_download_clip?: boolean

  do not download CLIP model even if it's not included in the checkpoint

• Optional dump_sysinfo?: boolean

  launch.py argument: dump limited sysinfo file (without information about extensions, options) to disk and quit

• Optional embeddings_dir?: string

  embeddings directory for textual inversion (default: embeddings)

• Optional enable_console_prompts?: boolean

  does not do anything

• Optional enable_insecure_extension_access?: boolean

  enable extensions tab regardless of other options

• Optional esrgan_models_path?: string

  Path to directory with ESRGAN model file(s).

• Optional f?: boolean

  ==SUPPRESS==

• Optional force_enable_xformers?: boolean

  enable xformers for cross attention layers regardless of whether the checking code thinks you can run it; do not make bug reports if this fails to work

• Optional freeze_settings?: boolean

  disable editing settings

• Optional gfpgan_dir?: string

  GFPGAN directory

• Optional gfpgan_model?: string

  GFPGAN model file name

• Optional gfpgan_models_path?: string

  Path to directory with GFPGAN model file(s).

• Optional gradio_allowed_path?: any[]

  add path to gradio's allowed_paths, make it possible to serve files from it

• Optional gradio_auth?: string

  set gradio authentication like "username:password"; or comma-delimit multiple like "u1:p1,u2:p2,u3:p3"

• Optional gradio_auth_path?: string

  set gradio authentication file path ex. "/path/to/auth/file" same auth format as --gradio-auth

• Optional gradio_debug?: boolean

  launch gradio with --debug option

• Optional gradio_img2img_tool?: string

  does not do anything

• Optional gradio_inpaint_tool?: string

  does not do anything

• Optional gradio_queue?: boolean

  does not do anything

• Optional hide_ui_dir_config?: boolean

  hide directory configuration from webui

• Optional hypernetwork_dir?: string

  hypernetwork directory

• Optional ldsr_models_path?: string

  Path to directory with LDSR model file(s).

• Optional listen?: boolean

  launch gradio with 0.0.0.0 as server name, allowing to respond to network requests

• Optional localizations_dir?: string

  localizations directory

• Optional log_startup?: boolean

  launch.py argument: print a detailed log of what's happening at startup

• Optional loglevel?: string

  log level; one of: CRITICAL, ERROR, WARNING, INFO, DEBUG

• Optional lora_dir?: string

  Path to directory with Lora networks.

• Optional lowram?: boolean

  load stable diffusion checkpoint weights to VRAM instead of RAM

• Optional lowvram?: boolean

  enable stable diffusion model optimizations for sacrificing a lot of speed for very low VRM usage

• Optional lyco_dir_backcompat?: string

  Path to directory with LyCORIS networks (for backawards compatibility; can also use --lyco-dir).

• Optional max_batch_count?: number

  maximum batch count value for the UI

• Optional medvram?: boolean

  enable stable diffusion model optimizations for sacrificing a little speed for low VRM usage

• Optional medvram_sdxl?: boolean

  enable --medvram optimization just for SDXL models

• Optional ngrok?: string

  ngrok authtoken, alternative to gradio --share

• Optional ngrok_options?: Record<string, any>

  The options to pass to ngrok in JSON format, e.g.: '{"authtoken_from_env":true, "basic_auth":"user:password", "oauth_provider":"google", "oauth_allow_emails":"user@asdf.com"}'

• Optional ngrok_region?: string

  does not do anything.

• Optional no_download_sd_model?: boolean

  don't download SD1.5 model even if no model is found in --ckpt-dir

• Optional no_gradio_queue?: boolean

  Disables gradio queue; causes the webpage to use http requests instead of websockets; was the default in earlier versions

• Optional no_half?: boolean

  do not switch the model to 16-bit floats

• Optional no_half_controlnet?: string

  do not switch the ControlNet models to 16-bit floats (only needed without --no-half)

• Optional no_half_vae?: boolean

  do not switch the VAE model to 16-bit floats

• Optional no_hashing?: boolean

  disable sha256 hashing of checkpoints to help loading performance

• Optional no_progressbar_hiding?: boolean

  do not hide progressbar in gradio UI (we hide it because it slows down ML if you have hardware acceleration in browser)

• Optional nowebui?: boolean

  use api=True to launch the API instead of the webui

• Optional opt_channelslast?: boolean

  change memory type for stable diffusion to channels last

• Optional opt_sdp_attention?: boolean

  prefer scaled dot product cross-attention layer optimization for automatic choice of optimization; requires PyTorch 2.*

• Optional opt_sdp_no_mem_attention?: boolean

  prefer scaled dot product cross-attention layer optimization without memory efficient attention for automatic choice of optimization, makes image generation deterministic; requires PyTorch 2.*

• Optional opt_split_attention?: boolean

  prefer Doggettx's cross-attention layer optimization for automatic choice of optimization

• Optional opt_split_attention_invokeai?: boolean

  prefer InvokeAI's cross-attention layer optimization for automatic choice of optimization

• Optional opt_split_attention_v1?: boolean

  prefer older version of split attention optimization for automatic choice of optimization

• Optional opt_sub_quad_attention?: boolean

  prefer memory efficient sub-quadratic cross-attention layer optimization for automatic choice of optimization

• Optional port?: string

  launch gradio with given server port, you need root/admin rights for ports < 1024, defaults to 7860 if available

• Optional precision?: string

  evaluate at this precision

• Optional realesrgan_models_path?: string

  Path to directory with RealESRGAN model file(s).

• Optional reinstall_torch?: boolean

  launch.py argument: install the appropriate version of torch even if you have some version already installed

• Optional reinstall_xformers?: boolean

  launch.py argument: install the appropriate version of xformers even if you have some version already installed

• Optional scunet_models_path?: string

  Path to directory with ScuNET model file(s).

• Optional server_name?: string

  Sets hostname of server

• Optional share?: boolean

  use share=True for gradio and make the UI accessible through their site

• Optional show_negative_prompt?: boolean

  does not do anything

• Optional skip_install?: boolean

  launch.py argument: skip installation of packages

• Optional skip_load_model_at_start?: boolean

  if load a model at web start, only take effect when --nowebui

• Optional skip_prepare_environment?: boolean

  launch.py argument: skip all environment preparation

• Optional skip_python_version_check?: boolean

  launch.py argument: do not check python version

• Optional skip_torch_cuda_test?: boolean

  launch.py argument: do not check if CUDA is able to work properly

• Optional skip_version_check?: boolean

  Do not check versions of torch and xformers

• Optional styles_file?: string

  filename to use for styles

• Optional sub_quad_chunk_threshold?: string

  the percentage of VRAM threshold for the sub-quadratic cross-attention layer optimization to use chunking

• Optional sub_quad_kv_chunk_size?: string

  kv chunk size for the sub-quadratic cross-attention layer optimization to use

• Optional sub_quad_q_chunk_size?: number

  query chunk size for the sub-quadratic cross-attention layer optimization to use

• Optional subpath?: string

  customize the subpath for gradio, use with reverse proxy

• Optional swinir_models_path?: string

  Path to directory with SwinIR model file(s).

• Optional test_server?: boolean

  launch.py argument: configure server for testing

• Optional textual_inversion_templates_dir?: string

  directory with textual inversion templates

• Optional theme?: string

  launches the UI with light or dark theme

• Optional timeout_keep_alive?: number

  set timeout_keep_alive for uvicorn

• Optional tls_certfile?: string

  Partially enables TLS, requires --tls-keyfile to fully function

• Optional tls_keyfile?: string

  Partially enables TLS, requires --tls-certfile to fully function

• Optional ui_config_file?: string

  filename to use for ui configuration

• Optional ui_debug_mode?: boolean

  Don't load model to quickly launch UI

• Optional ui_settings_file?: string

  filename to use for ui settings

• Optional unload_gfpgan?: boolean

  does not do anything.

• Optional upcast_sampling?: boolean

  upcast sampling. No effect with --no-half. Usually produces similar results to --no-half with better performance while using less memory.

• Optional update_all_extensions?: boolean

  launch.py argument: download updates for all extensions when starting the program

• Optional update_check?: boolean

  launch.py argument: check for updates at startup

• Optional use_cpu?: any[]

  use CPU as torch device for specified modules

• Optional use_ipex?: boolean

  use Intel XPU as torch device

• Optional use_textbox_seed?: boolean

  use textbox for seeds in UI (no up/down, but possible to input long seeds)

• Optional vae_dir?: string

  Path to directory with VAE files

• Optional vae_path?: string

  Checkpoint to use as VAE; setting this argument disables all settings related to VAE

• Optional xformers?: boolean

  enable xformers for cross attention layers

• Optional xformers_flash_attention?: boolean

  enable xformers with Flash Attention to improve reproducibility (supported for SD2.x or variant only)