InvokeAI v2.3.4 --help output

invokeAiHelp.txt

usage: invokeai [-h] [--laion400m LAION400M] [--weights WEIGHTS] [--version]
                [--root_dir ROOT_DIR] [--config CONF] [--model MODEL]
                [--weight_dirs WEIGHT_DIRS [WEIGHT_DIRS ...]]
                [--png_compression {0,1,2,3,4,5,6,7,8}] [-F]
                [--max_loaded_models MAX_LOADED_MODELS] [--free_gpu_mem]
                [--sequential_guidance] [--xformers | --no-xformers]
                [--always_use_cpu] [--precision PRECISION]
                [--ckpt_convert | --no-ckpt_convert]
                [--internet | --no-internet]
                [--nsfw_checker | --no-nsfw_checker | --safety_checker | --no-safety_checker]
                [--autoimport AUTOIMPORT] [--autoconvert AUTOCONVERT]
                [--patchmatch | --no-patchmatch] [--from_file INFILE]
                [--outdir OUTDIR] [--prompt_as_dir] [--fnformat FNFORMAT]
                [-s STEPS] [-W WIDTH] [-H HEIGHT] [-C CFG_SCALE]
                [--sampler SAMPLER_NAME] [--log_tokenization] [-f STRENGTH]
                [-T | -fit | --fit | --no-fit] [--grid | --no-grid | -g]
                [--embedding_directory EMBEDDING_PATH]
                [--lora_directory LORA_PATH] [--embeddings | --no-embeddings]
                [--enable_image_debugging] [--karras_max KARRAS_MAX]
                [--no_restore] [--no_upscale]
                [--esrgan_bg_tile ESRGAN_BG_TILE]
                [--esrgan_denoise_str ESRGAN_DENOISE_STR]
                [--gfpgan_model_path GFPGAN_MODEL_PATH] [--web]
                [--web_develop] [--web_verbose] [--cors [CORS ...]]
                [--host HOST] [--port PORT] [--certfile CERTFILE]
                [--keyfile KEYFILE] [--gui]

Generate images using Stable Diffusion. Use --web to launch the web interface.
Use --from_file to load prompts from a file path or standard input ("-").
Otherwise you will be dropped into an interactive command prompt (type -h for
help.) Other command-line arguments are defaults that can usually be
overridden prompt the command prompt.

options:
  -h, --help            show this help message and exit

General:
  --version, -V         Print InvokeAI version number

Model selection:
  --root_dir ROOT_DIR   Path to directory containing "models", "outputs" and
                        "configs". If not present will read from environment
                        variable INVOKEAI_ROOT. Defaults to ~/invokeai.
  --config CONF, -c CONF, -config CONF
                        Path to configuration file for alternate models.
  --model MODEL         Indicates which diffusion model to load (defaults to
                        "default" stanza in configs/models.yaml)
  --weight_dirs WEIGHT_DIRS [WEIGHT_DIRS ...]
                        List of one or more directories that will be auto-
                        scanned for new model weights to import
  --png_compression {0,1,2,3,4,5,6,7,8}, -z {0,1,2,3,4,5,6,7,8}
                        level of PNG compression, from 0 (none) to 9
                        (maximum). Default is 6.
  -F, --full_precision  Deprecated way to set --precision=float32
  --max_loaded_models MAX_LOADED_MODELS
                        Maximum number of models to keep in memory for fast
                        switching, including the one in GPU
  --free_gpu_mem        Force free gpu memory before final decoding
  --sequential_guidance
                        Calculate guidance in serial instead of in parallel,
                        lowering memory requirement at the expense of speed
  --xformers, --no-xformers
                        Enable/disable xformers support (default enabled if
                        installed) (default: True)
  --always_use_cpu      Force use of CPU even if GPU is available
  --precision PRECISION
                        Set model precision. Defaults to auto selected based
                        on device. Options: auto, float32, autocast, float16
  --ckpt_convert, --no-ckpt_convert
                        Deprecated option. Legacy ckpt files are now always
                        converted to diffusers when loaded. (default: True)
  --internet, --no-internet
                        Indicate whether internet is available for just-in-
                        time model downloading (default: probe automatically).
                        (default: True)
  --nsfw_checker, --no-nsfw_checker, --safety_checker, --no-safety_checker
                        Check for and blur potentially NSFW images. Use --no-
                        nsfw_checker to disable. (default: False)
  --autoimport AUTOIMPORT
                        Check the indicated directory for .ckpt/.safetensors
                        weights files at startup and import directly
  --autoconvert AUTOCONVERT
                        Check the indicated directory for .ckpt/.safetensors
                        weights files at startup and import as optimized
                        diffuser models
  --patchmatch, --no-patchmatch
                        Load the patchmatch extension for outpainting. Use
                        --no-patchmatch to disable. (default: True)

Input/output:
  --from_file INFILE    If specified, load prompts from this file
  --outdir OUTDIR, -o OUTDIR
                        Directory to save generated images and a log of
                        prompts and seeds. Default: ROOTDIR/outputs
  --prompt_as_dir, -p   Place images in subdirectories named after the prompt.

Web server:
  --web                 Start in web server mode.
  --web_develop         Start in web server development mode.
  --web_verbose         Enables verbose logging
  --cors [CORS ...]     Additional allowed origins, comma-separated
  --host HOST           Web server: Host or IP to listen on. Set to 0.0.0.0 to
                        accept traffic from other devices on your network.
  --port PORT           Web server: Port to listen on
  --certfile CERTFILE   Web server: Path to certificate file to use for SSL.
                        Use together with --keyfile
  --keyfile KEYFILE     Web server: Path to private key file to use for SSL.
                        Use together with --certfile
  --gui                 Start InvokeAI GUI

Rendering:
  --fnformat FNFORMAT   Overwrite the filename format. You can use any
                        argument as wildcard enclosed in curly braces. Default
                        is {prefix}.{seed}.png
  -s STEPS, --steps STEPS
                        Number of steps
  -W WIDTH, --width WIDTH
                        Image width, multiple of 64
  -H HEIGHT, --height HEIGHT
                        Image height, multiple of 64
  -C CFG_SCALE, --cfg_scale CFG_SCALE
                        Classifier free guidance (CFG) scale - higher numbers
                        cause generator to "try" harder.
  --sampler SAMPLER_NAME, -A SAMPLER_NAME, -m SAMPLER_NAME
                        Set the default sampler. Supported samplers: ddim,
                        k_dpm_2_a, k_dpm_2, k_dpmpp_2_a, k_dpmpp_2, k_euler_a,
                        k_euler, k_heun, k_lms, plms, pndm
  --log_tokenization, -t
                        shows how the prompt is split into tokens
  -f STRENGTH, --strength STRENGTH
                        img2img strength for noising/unnoising. 0.0 preserves
                        image exactly, 1.0 replaces it completely
  -T, -fit, --fit, --no-fit
                        If specified, will resize the input image to fit
                        within the dimensions of width x height (512x512
                        default)
  --grid, --no-grid, -g
                        generate a grid
  --embedding_directory EMBEDDING_PATH, --embedding_path EMBEDDING_PATH
                        Path to a directory containing .bin and/or .pt files,
                        or a single .bin/.pt file. You may use subdirectories.
                        (default is ROOTDIR/embeddings)
  --lora_directory LORA_PATH
                        Path to a directory containing LoRA files;
                        subdirectories are not supported. (default is
                        ROOTDIR/loras)
  --embeddings, --no-embeddings
                        Enable embedding directory (default). Use --no-
                        embeddings to disable. (default: True)
  --enable_image_debugging
                        Generates debugging image to display
  --karras_max KARRAS_MAX
                        control the point at which the K* samplers will shift
                        from using the Karras noise schedule (good for low
                        step counts) to the LatentDiffusion noise schedule
                        (good for high step counts). Set to 0 to use
                        LatentDiffusion for all step values, and to a high
                        value (e.g. 1000) to use Karras for all step values.
                        [29].

Postprocessing:
  --no_restore          Disable face restoration with GFPGAN or codeformer
  --no_upscale          Disable upscaling with ESRGAN
  --esrgan_bg_tile ESRGAN_BG_TILE
                        Tile size for background sampler, 0 for no tile during
                        testing. Default: 400.
  --esrgan_denoise_str ESRGAN_DENOISE_STR
                        esrgan denoise str. 0 is no denoise, 1 is max denoise.
                        Default: 0.75
  --gfpgan_model_path GFPGAN_MODEL_PATH
                        Indicates the path to the GFPGAN model

Deprecated options:
  --laion400m LAION400M
  --weights WEIGHTS