Overview — dgenerate 3.7.1 documentation (2024)

usage: dgenerate [-h] [-v] [--version] [--file | --shell | --no-stdin | --console] [--plugin-modules PATH [PATH ...]] [--sub-command SUB_COMMAND] [--sub-command-help [SUB_COMMAND ...]] [-ofm] [--templates-help [VARIABLE_NAME ...]] [--directives-help [DIRECTIVE_NAME ...]] [--functions-help [FUNCTION_NAME ...]] [-mt MODEL_TYPE] [-rev BRANCH] [-var VARIANT] [-sbf SUBFOLDER] [-atk TOKEN] [-bs INTEGER] [-bgs SIZE] [-un UNET_URI] [-un2 UNET_URI] [-vae VAE_URI] [-vt] [-vs] [-lra LORA_URI [LORA_URI ...]] [-ti URI [URI ...]] [-cn CONTROL_NET_URI [CONTROL_NET_URI ...]] [-sch SCHEDULER_URI] [-mqo | -mco] [--s-cascade-decoder MODEL_URI] [-dqo] [-dco] [--s-cascade-decoder-prompts PROMPT [PROMPT ...]] [--s-cascade-decoder-inference-steps INTEGER [INTEGER ...]] [--s-cascade-decoder-guidance-scales INTEGER [INTEGER ...]] [--s-cascade-decoder-scheduler SCHEDULER_URI] [--sdxl-refiner MODEL_URI] [-rqo] [-rco] [--sdxl-refiner-scheduler SCHEDULER_URI] [--sdxl-refiner-edit] [--sdxl-second-prompts PROMPT [PROMPT ...]] [--sdxl-aesthetic-scores FLOAT [FLOAT ...]] [--sdxl-crops-coords-top-left COORD [COORD ...]] [--sdxl-original-size SIZE [SIZE ...]] [--sdxl-target-size SIZE [SIZE ...]] [--sdxl-negative-aesthetic-scores FLOAT [FLOAT ...]] [--sdxl-negative-original-sizes SIZE [SIZE ...]] [--sdxl-negative-target-sizes SIZE [SIZE ...]] [--sdxl-negative-crops-coords-top-left COORD [COORD ...]] [--sdxl-refiner-prompts PROMPT [PROMPT ...]] [--sdxl-refiner-clip-skips INTEGER [INTEGER ...]] [--sdxl-refiner-second-prompts PROMPT [PROMPT ...]] [--sdxl-refiner-aesthetic-scores FLOAT [FLOAT ...]] [--sdxl-refiner-crops-coords-top-left COORD [COORD ...]] [--sdxl-refiner-original-sizes SIZE [SIZE ...]] [--sdxl-refiner-target-sizes SIZE [SIZE ...]] [--sdxl-refiner-negative-aesthetic-scores FLOAT [FLOAT ...]] [--sdxl-refiner-negative-original-sizes SIZE [SIZE ...]] [--sdxl-refiner-negative-target-sizes SIZE [SIZE ...]] [--sdxl-refiner-negative-crops-coords-top-left COORD [COORD ...]] [-hnf FLOAT [FLOAT ...]] [-ri INT [INT ...]] [-rg FLOAT [FLOAT ...]] [-rgr FLOAT [FLOAT ...]] [-sc] [-d DEVICE] [-t DTYPE] [-s SIZE] [-na] [-o PATH] [-op PREFIX] [-ox] [-oc] [-om] [-p PROMPT [PROMPT ...]] [-cs INTEGER [INTEGER ...]] [-se SEED [SEED ...]] [-sei] [-gse COUNT] [-af FORMAT] [-if FORMAT] [-nf] [-fs FRAME_NUMBER] [-fe FRAME_NUMBER] [-is SEED [SEED ...]] [-sip PROCESSOR_URI [PROCESSOR_URI ...]] [-mip PROCESSOR_URI [PROCESSOR_URI ...]] [-cip PROCESSOR_URI [PROCESSOR_URI ...]] [--image-processor-help [PROCESSOR_NAME ...]] [-pp PROCESSOR_URI [PROCESSOR_URI ...]] [-iss FLOAT [FLOAT ...] | -uns INTEGER [INTEGER ...]] [-gs FLOAT [FLOAT ...]] [-igs FLOAT [FLOAT ...]] [-gr FLOAT [FLOAT ...]] [-ifs INTEGER [INTEGER ...]] [-mc EXPR [EXPR ...]] [-pmc EXPR [EXPR ...]] [-umc EXPR [EXPR ...]] [-vmc EXPR [EXPR ...]] [-cmc EXPR [EXPR ...]] model_pathBatch image generation and manipulation tool supporting Stable Diffusion and related techniques /algorithms, with support for video and animated image processing.positional arguments: model_path huggingface model repository slug, huggingface blob link to a model file, path to folder on disk, or path to a .pt, .pth, .bin, .ckpt, or .safetensors file.options: -h, --help show this help message and exit -v, --verbose Output information useful for debugging, such as pipeline call and model load parameters. --version Show dgenerate's version and exit --file Convenience argument for reading a configuration script from a file instead of using a pipe. This is a meta argument which can not be used within a configuration script and is only valid from the command line or during a popen invocation of dgenerate. --shell When reading configuration from STDIN (a pipe), read forever, even when configuration errors occur. This allows dgenerate to run in the background and be communicated with by another process sending it commands. Launching dgenerate with this option and not piping it input will attach it to the terminal like a shell. Entering configuration into this shell will require two newlines to submit a command due to parsing lookahead. IE: two presses of the enter key. This is a meta argument which can not be used within a configuration script and is only valid from the command line or during a popen invocation of dgenerate. --no-stdin Can be used to indicate to dgenerate that it will not receive any piped in input. This is useful for running dgenerate via popen from python or another application using normal arguments, where it would otherwise try to read from STDIN and block forever because it is not attached to a terminal. This is a meta argument which can not be used within a configuration script and is only valid from the command line or during a popen invocation of dgenerate. --console Launch a terminal-like tkinter GUI that communicates with an instance of dgenerate running in the background. This allows you to interactively write dgenerate config scripts as if dgenerate were a shell / REPL. This is a meta argument which can not be used within a configuration script and is only valid from the command line or during a popen invocation of dgenerate. --plugin-modules PATH [PATH ...] Specify one or more plugin module folder paths (folder containing __init__.py) or python .py file paths, or python module names to load as plugins. Plugin modules can currently implement image processors and config directives. --sub-command SUB_COMMAND Specify the name a sub-command to invoke. dgenerate exposes some extra image processing functionality through the use of sub-commands. Sub commands essentially replace the entire set of accepted arguments with those of a sub-command which implements additional functionality. See --sub-command-help for a list of sub-commands and help. --sub-command-help [SUB_COMMAND ...] List available sub-commands, providing sub-command names will produce their documentation. Calling a subcommand with "--sub-command name --help" will produce argument help output for that subcommand. -ofm, --offline-mode Whether dgenerate should try to download huggingface models that do not exist in the disk cache, or only use what is available in the cache. Referencing a model on huggingface that has not been cached because it was not previously downloaded will result in a failure when using this option. --templates-help [VARIABLE_NAME ...] Print a list of template variables available in the interpreter environment used for dgenerate config scripts, particularly the variables set after a dgenerate invocation occurs. When used as a command line option, their values are not presented, just their names and types. Specifying names will print type information for those variable names. --directives-help [DIRECTIVE_NAME ...] Print a list of directives available in the interpreter environment used for dgenerate config scripts. Providing names will print documentation for the specified directive names. When used with --plugin-modules, directives implemented by the specified plugins will also be listed. --functions-help [FUNCTION_NAME ...] Print a list of template functions available in the interpreter environment used for dgenerate config scripts. Providing names will print documentation for the specified function names. When used with --plugin- modules, functions implemented by the specified plugins will also be listed. -mt MODEL_TYPE, --model-type MODEL_TYPE Use when loading different model types. Currently supported: torch, torch- pix2pix, torch-sdxl, torch-sdxl-pix2pix, torch-upscaler-x2, torch- upscaler-x4, torch-if, torch-ifs, torch-ifs-img2img, or torch-s-cascade. (default: torch) -rev BRANCH, --revision BRANCH The model revision to use when loading from a huggingface repository, (The git branch / tag, default is "main") -var VARIANT, --variant VARIANT If specified when loading from a huggingface repository or folder, load weights from "variant" filename, e.g. "pytorch_model.<variant>.safetensors". Defaults to automatic selection. This option is ignored if using flax. -sbf SUBFOLDER, --subfolder SUBFOLDER Main model subfolder. If specified when loading from a huggingface repository or folder, load weights from the specified subfolder. -atk TOKEN, --auth-token TOKEN Huggingface auth token. Required to download restricted repositories that have access permissions granted to your huggingface account. -bs INTEGER, --batch-size INTEGER The number of image variations to produce per set of individual diffusion parameters in one rendering step simultaneously on a single GPU. When using flax, batch size is controlled by the environmental variable CUDA_VISIBLE_DEVICES which is a comma separated list of GPU device numbers (as listed by nvidia-smi). Usage of this argument with --model-type flax* will cause an error, diffusion with flax will generate an image on every GPU that is visible to CUDA and this is currently unchangeable. When generating animations with a --batch-size greater than one, a separate animation (with the filename suffix "animation_N") will be written to for each image in the batch. If --batch-grid-size is specified when producing an animation then the image grid is used for the output frames. During animation rendering each image in the batch will still be written to the output directory along side the produced animation as either suffixed files or image grids depending on the options you choose. (Torch Default:1) -bgs SIZE, --batch-grid-size SIZE Produce a single image containing a grid of images with the number of COLUMNSxROWS given to this argument when --batch-size is greater than 1, or when using flax with multiple GPUs visible (via the environmental variable CUDA_VISIBLE_DEVICES). If not specified with a --batch-size greater than 1, images will be written individually with an image number suffix (image_N) in the filename signifying which image in the batch they are. -un UNET_URI, --unet UNET_URI Specify a UNet using a URI. Examples: "huggingface/unet", "huggingface/unet;revision=main", "unet_folder_on_disk". Blob links / single file loads are not supported for UNets. The "revision" argument specifies the model revision to use for the UNet when loading from huggingface repository or blob link, (The git branch / tag, default is "main"). The "variant" argument specifies the UNet model variant, it is only supported for torch type models it is not supported for flax. If "variant" is specified when loading from a huggingface repository or folder, weights will be loaded from "variant" filename, e.g. "pytorch_model.<variant>.safetensors. "variant" defaults to the value of --variant if it is not specified in the URI. The "subfolder" argument specifies the UNet model subfolder, if specified when loading from a huggingface repository or folder, weights from the specified subfolder. The "dtype" argument specifies the UNet model precision, it defaults to the value of -t/--dtype and should be one of: auto, bfloat16, float16, or float32. If you wish to load weights directly from a path on disk, you must point this argument at the folder they exist in, which should also contain the config.json file for the UNet. For example, a downloaded repository folder from huggingface. -un2 UNET_URI, --unet2 UNET_URI Specify a second UNet, this is only valid when using SDXL or Stable Cascade model types. This UNet will be used for the SDXL refiner, or Stable Cascade decoder model. -vae VAE_URI, --vae VAE_URI Specify a VAE using a URI. When using torch models the URI syntax is: "AutoEncoderClass;model=(huggingface repository slug/blob link or file/folder path)". Examples: "AutoencoderKL;model=vae.pt", "AsymmetricAutoencoderKL;model=huggingface/vae", "AutoencoderTiny;model=huggingface/vae", "ConsistencyDecoderVAE;model=huggingface/vae". When using a Flax model, there is currently only one available encoder class: "FlaxAutoencoderKL;model=huggingface/vae". The AutoencoderKL encoder class accepts huggingface repository slugs/blob links, .pt, .pth, .bin, .ckpt, and .safetensors files. Other encoders can only accept huggingface repository slugs/blob links, or a path to a folder on disk with the model configuration and model file(s). If an AutoencoderKL VAE model file exists at a URL which serves the file as a raw download, you may provide an http/https link to it and it will be downloaded to dgenerates web cache. Aside from the "model" argument, there are four other optional arguments that can be specified, these include "revision", "variant", "subfolder", "dtype". They can be specified as so in any order, they are not positional: "AutoencoderKL;model=huggingface/vae;revision=main;variant=fp1 6;subfolder=sub_folder;dtype=float16". The "revision" argument specifies the model revision to use for the VAE when loading from huggingface repository or blob link, (The git branch / tag, default is "main"). The "variant" argument specifies the VAE model variant, it is only supported for torch type models it is not supported for flax. If "variant" is specified when loading from a huggingface repository or folder, weights will be loaded from "variant" filename, e.g. "pytorch_model.<variant>.safetensors. "variant" in the case of --vae does not default to the value of --variant to prevent failures during common use cases. The "subfolder" argument specifies the VAE model subfolder, if specified when loading from a huggingface repository or folder, weights from the specified subfolder. The "dtype" argument specifies the VAE model precision, it defaults to the value of -t/--dtype and should be one of: auto, bfloat16, float16, or float32. If you wish to load a weights file directly from disk, the simplest way is: --vae "AutoencoderKL;my_vae.safetensors", or with a dtype "AutoencoderKL;my_vae.safetensors;dtype=float16". All loading arguments except "dtype" are unused in this case and may produce an error message if used. If you wish to load a specific weight file from a huggingface repository, use the blob link loading syntax: --vae "AutoencoderKL;https://huggingface.co/UserName/repository- name/blob/main/vae_model.safetensors", the "revision" argument may be used with this syntax. -vt, --vae-tiling Enable VAE tiling (torch Stable Diffusion only). Assists in the generation of large images with lower memory overhead. The VAE will split the input tensor into tiles to compute decoding and encoding in several steps. This is useful for saving a large amount of memory and to allow processing larger images. Note that if you are using --control-nets you may still run into memory issues generating large images, or with --batch-size greater than 1. -vs, --vae-slicing Enable VAE slicing (torch Stable Diffusion models only). Assists in the generation of large images with lower memory overhead. The VAE will split the input tensor in slices to compute decoding in several steps. This is useful to save some memory, especially when --batch-size is greater than 1. Note that if you are using --control-nets you may still run into memory issues generating large images. -lra LORA_URI [LORA_URI ...], --loras LORA_URI [LORA_URI ...] Specify one or more LoRA models using URIs (flax not supported). These should be a huggingface repository slug, path to model file on disk (for example, a .pt, .pth, .bin, .ckpt, or .safetensors file), or model folder containing model files. If a LoRA model file exists at a URL which serves the file as a raw download, you may provide an http/https link to it and it will be downloaded to dgenerates web cache. huggingface blob links are not supported, see "subfolder" and "weight-name" below instead. Optional arguments can be provided after a LoRA model specification, these include: "scale", "revision", "subfolder", and "weight-name". They can be specified as so in any order, they are not positional: "huggingface/lora;scale=1.0;revision=main;subfolder=repo_subfolder;weight- name=lora.safetensors". The "scale" argument indicates the scale factor of the LoRA. The "revision" argument specifies the model revision to use for the LoRA when loading from huggingface repository, (The git branch / tag, default is "main"). The "subfolder" argument specifies the LoRA model subfolder, if specified when loading from a huggingface repository or folder, weights from the specified subfolder. The "weight-name" argument indicates the name of the weights file to be loaded when loading from a huggingface repository or folder on disk. If you wish to load a weights file directly from disk, the simplest way is: --loras "my_lora.safetensors", or with a scale "my_lora.safetensors;scale=1.0", all other loading arguments are unused in this case and may produce an error message if used. -ti URI [URI ...], --textual-inversions URI [URI ...] Specify one or more Textual Inversion models using URIs (flax and SDXL not supported). These should be a huggingface repository slug, path to model file on disk (for example, a .pt, .pth, .bin, .ckpt, or .safetensors file), or model folder containing model files. If a Textual Inversion model file exists at a URL which serves the file as a raw download, you may provide an http/https link to it and it will be downloaded to dgenerates web cache. huggingface blob links are not supported, see "subfolder" and "weight-name" below instead. Optional arguments can be provided after the Textual Inversion model specification, these include: "token", "revision", "subfolder", and "weight-name". They can be specified as so in any order, they are not positional: "huggingface/ti_model;revision=main;subfolder=repo_subfolder;weight- name=ti_model.safetensors". The "token" argument can be used to override the prompt token used for the textual inversion prompt embedding. For normal Stable Diffusion the default token value is provided by the model itself, but for Stable Diffusion XL the default token value is equal to the model file name with no extension and all spaces replaced by underscores. The "revision" argument specifies the model revision to use for the Textual Inversion model when loading from huggingface repository, (The git branch / tag, default is "main"). The "subfolder" argument specifies the Textual Inversion model subfolder, if specified when loading from a huggingface repository or folder, weights from the specified subfolder. The "weight-name" argument indicates the name of the weights file to be loaded when loading from a huggingface repository or folder on disk. If you wish to load a weights file directly from disk, the simplest way is: --textual-inversions "my_ti_model.safetensors", all other loading arguments are unused in this case and may produce an error message if used. -cn CONTROL_NET_URI [CONTROL_NET_URI ...], --control-nets CONTROL_NET_URI [CONTROL_NET_URI ...] Specify one or more ControlNet models using URIs. This should be a huggingface repository slug / blob link, path to model file on disk (for example, a .pt, .pth, .bin, .ckpt, or .safetensors file), or model folder containing model files. If a ControlNet model file exists at a URL which serves the file as a raw download, you may provide an http/https link to it and it will be downloaded to dgenerates web cache. Optional arguments can be provided after the ControlNet model specification, for torch these include: "scale", "start", "end", "revision", "variant", "subfolder", and "dtype". For flax: "scale", "revision", "subfolder", "dtype", "from_torch" (bool) They can be specified as so in any order, they are not positional: "huggingface/controlnet;scale=1.0;start=0.0;end=1.0;revision=main;variant= fp16;subfolder=repo_subfolder;dtype=float16". The "scale" argument specifies the scaling factor applied to the ControlNet model, the default value is 1.0. The "start" (only for --model-type "torch*") argument specifies at what fraction of the total inference steps to begin applying the ControlNet, defaults to 0.0, IE: the very beginning. The "end" (only for --model-type "torch*") argument specifies at what fraction of the total inference steps to stop applying the ControlNet, defaults to 1.0, IE: the very end. The "revision" argument specifies the model revision to use for the ControlNet model when loading from huggingface repository, (The git branch / tag, default is "main"). The "variant" (only for --model-type "torch*") argument specifies the ControlNet model variant, if "variant" is specified when loading from a huggingface repository or folder, weights will be loaded from "variant" filename, e.g. "pytorch_model.<variant>.safetensors. "variant" defaults to automatic selection and is ignored if using flax. "variant" in the case of --control-nets does not default to the value of --variant to prevent failures during common use cases. The "subfolder" argument specifies the ControlNet model subfolder, if specified when loading from a huggingface repository or folder, weights from the specified subfolder. The "dtype" argument specifies the ControlNet model precision, it defaults to the value of -t/--dtype and should be one of: auto, bfloat16, float16, or float32. The "from_torch" (only for --model-type flax) this argument specifies that the ControlNet is to be loaded and converted from a huggingface repository or file that is designed for pytorch. (Defaults to false) If you wish to load a weights file directly from disk, the simplest way is: --control-nets "my_controlnet.safetensors" or --control-nets "my_controlnet.safetensors;scale=1.0;dtype=float16", all other loading arguments aside from "scale" and "dtype" are unused in this case and may produce an error message if used ("from_torch" is available when using flax). If you wish to load a specific weight file from a huggingface repository, use the blob link loading syntax: --control-nets "https://huggingface.co/UserName/repository- name/blob/main/controlnet.safetensors", the "revision" argument may be used with this syntax. -sch SCHEDULER_URI, --scheduler SCHEDULER_URI Specify a scheduler (sampler) by URI. Passing "help" to this argument will print the compatible schedulers for a model without generating any images. Passing "helpargs" will yield a help message with a list of overridable arguments for each scheduler and their typical defaults. Arguments listed by "helpargs" can be overridden using the URI syntax typical to other dgenerate URI arguments. Torch schedulers: (DDIMScheduler, DDPMScheduler, PNDMScheduler, LMSDiscreteScheduler, EulerDiscreteScheduler, HeunDiscreteScheduler, EulerAncestralDiscreteScheduler, DPMSolverMultistepScheduler, DPMSolverSinglestepScheduler, KDPM2DiscreteScheduler, KDPM2AncestralDiscreteScheduler, DEISMultistepScheduler, UniPCMultistepScheduler, DPMSolverSDEScheduler, EDMEulerScheduler). -mqo, --model-sequential-offload Force sequential model offloading for the main pipeline, this may drastically reduce memory consumption and allow large models to run when they would otherwise not fit in your GPUs VRAM. Inference will be much slower. Mutually exclusive with --model-cpu-offload -mco, --model-cpu-offload Force model cpu offloading for the main pipeline, this may reduce memory consumption and allow large models to run when they would otherwise not fit in your GPUs VRAM. Inference will be slower. Mutually exclusive with --model-sequential-offload --s-cascade-decoder MODEL_URI Specify a Stable Cascade (torch-s-cascade) decoder model path using a URI. This should be a huggingface repository slug / blob link, path to model file on disk (for example, a .pt, .pth, .bin, .ckpt, or .safetensors file), or model folder containing model files. Optional arguments can be provided after the decoder model specification, these include: "revision", "variant", "subfolder", and "dtype". They can be specified as so in any order, they are not positional: "huggingface/decoder_model;revision=main;v ariant=fp16;subfolder=repo_subfolder;dtype=float16". The "revision" argument specifies the model revision to use for the Textual Inversion model when loading from huggingface repository, (The git branch / tag, default is "main"). The "variant" argument specifies the decoder model variant and defaults to the value of --variant. When "variant" is specified when loading from a huggingface repository or folder, weights will be loaded from "variant" filename, e.g. "pytorch_model.<variant>.safetensors. The "subfolder" argument specifies the decoder model subfolder, if specified when loading from a huggingface repository or folder, weights from the specified subfolder. The "dtype" argument specifies the Stable Cascade decoder model precision, it defaults to the value of -t/--dtype and should be one of: auto, bfloat16, float16, or float32. If you wish to load a weights file directly from disk, the simplest way is: --sdxl-refiner "my_decoder.safetensors" or --sdxl-refiner "my_decoder.safetensors;dtype=float16", all other loading arguments aside from "dtype" are unused in this case and may produce an error message if used. If you wish to load a specific weight file from a huggingface repository, use the blob link loading syntax: --s-cascade-decoder "https://huggingface.co/UserName/repository- name/blob/main/decoder.safetensors", the "revision" argument may be used with this syntax. -dqo, --s-cascade-decoder-sequential-offload Force sequential model offloading for the Stable Cascade decoder pipeline, this may drastically reduce memory consumption and allow large models to run when they would otherwise not fit in your GPUs VRAM. Inference will be much slower. Mutually exclusive with --s-cascade-decoder-cpu-offload -dco, --s-cascade-decoder-cpu-offload Force model cpu offloading for the Stable Cascade decoder pipeline, this may reduce memory consumption and allow large models to run when they would otherwise not fit in your GPUs VRAM. Inference will be slower. Mutually exclusive with --s-cascade-decoder-sequential-offload --s-cascade-decoder-prompts PROMPT [PROMPT ...] One or more prompts to try with the Stable Cascade decoder model, by default the decoder model gets the primary prompt, this argument overrides that with a prompt of your choosing. The negative prompt component can be specified with the same syntax as --prompts --s-cascade-decoder-inference-steps INTEGER [INTEGER ...] One or more inference steps values to try with the Stable Cascade decoder. (default: [10]) --s-cascade-decoder-guidance-scales INTEGER [INTEGER ...] One or more guidance scale values to try with the Stable Cascade decoder. (default: [0]) --s-cascade-decoder-scheduler SCHEDULER_URI Specify a scheduler (sampler) by URI for the Stable Cascade decoder pass. Operates the exact same way as --scheduler including the "help" option. Passing 'helpargs' will yield a help message with a list of overridable arguments for each scheduler and their typical defaults. Defaults to the value of --scheduler. --sdxl-refiner MODEL_URI Specify a Stable Diffusion XL (torch-sdxl) refiner model path using a URI. This should be a huggingface repository slug / blob link, path to model file on disk (for example, a .pt, .pth, .bin, .ckpt, or .safetensors file), or model folder containing model files. Optional arguments can be provided after the SDXL refiner model specification, these include: "revision", "variant", "subfolder", and "dtype". They can be specified as so in any order, they are not positional: "huggingface/refiner_model_xl;re vision=main;variant=fp16;subfolder=repo_subfolder;dtype=float16". The "revision" argument specifies the model revision to use for the Textual Inversion model when loading from huggingface repository, (The git branch / tag, default is "main"). The "variant" argument specifies the SDXL refiner model variant and defaults to the value of --variant. When "variant" is specified when loading from a huggingface repository or folder, weights will be loaded from "variant" filename, e.g. "pytorch_model.<variant>.safetensors. The "subfolder" argument specifies the SDXL refiner model subfolder, if specified when loading from a huggingface repository or folder, weights from the specified subfolder. The "dtype" argument specifies the SDXL refiner model precision, it defaults to the value of -t/--dtype and should be one of: auto, bfloat16, float16, or float32. If you wish to load a weights file directly from disk, the simplest way is: --sdxl-refiner "my_sdxl_refiner.safetensors" or --sdxl-refiner "my_sdxl_refiner.safetensors;dtype=float16", all other loading arguments aside from "dtype" are unused in this case and may produce an error message if used. If you wish to load a specific weight file from a huggingface repository, use the blob link loading syntax: --sdxl-refiner "https://huggingface.co/UserName/repository- name/blob/main/refiner_model.safetensors", the "revision" argument may be used with this syntax. -rqo, --sdxl-refiner-sequential-offload Force sequential model offloading for the SDXL refiner pipeline, this may drastically reduce memory consumption and allow large models to run when they would otherwise not fit in your GPUs VRAM. Inference will be much slower. Mutually exclusive with --refiner-cpu-offload -rco, --sdxl-refiner-cpu-offload Force model cpu offloading for the SDXL refiner pipeline, this may reduce memory consumption and allow large models to run when they would otherwise not fit in your GPUs VRAM. Inference will be slower. Mutually exclusive with --refiner-sequential-offload --sdxl-refiner-scheduler SCHEDULER_URI Specify a scheduler (sampler) by URI for the SDXL refiner pass. Operates the exact same way as --scheduler including the "help" option. Passing 'helpargs' will yield a help message with a list of overridable arguments for each scheduler and their typical defaults. Defaults to the value of --scheduler. --sdxl-refiner-edit Force the SDXL refiner to operate in edit mode instead of cooperative denoising mode as it would normally do for inpainting and ControlNet usage. The main model will preform the full amount of inference steps requested by --inference-steps. The output of the main model will be passed to the refiner model and processed with an image seed strength in img2img mode determined by (1.0 - high-noise-fraction) --sdxl-second-prompts PROMPT [PROMPT ...] One or more secondary prompts to try using SDXL's secondary text encoder. By default the model is passed the primary prompt for this value, this option allows you to choose a different prompt. The negative prompt component can be specified with the same syntax as --prompts --sdxl-aesthetic-scores FLOAT [FLOAT ...] One or more Stable Diffusion XL (torch-sdxl) "aesthetic-score" micro- conditioning parameters. Used to simulate an aesthetic score of the generated image by influencing the positive text condition. Part of SDXL's micro-conditioning as explained in section 2.2 of [https://huggingface.co/papers/2307.01952]. --sdxl-crops-coords-top-left COORD [COORD ...] One or more Stable Diffusion XL (torch-sdxl) "negative-crops-coords-top- left" micro-conditioning parameters in the format "0,0". --sdxl-crops- coords-top-left can be used to generate an image that appears to be "cropped" from the position --sdxl-crops-coords-top-left downwards. Favorable, well-centered images are usually achieved by setting --sdxl- crops-coords-top-left to "0,0". Part of SDXL's micro-conditioning as explained in section 2.2 of [https://huggingface.co/papers/2307.01952]. --sdxl-original-size SIZE [SIZE ...], --sdxl-original-sizes SIZE [SIZE ...] One or more Stable Diffusion XL (torch-sdxl) "original-size" micro- conditioning parameters in the format (WIDTH)x(HEIGHT). If not the same as --sdxl-target-size the image will appear to be down or up-sampled. --sdxl- original-size defaults to --output-size or the size of any input images if not specified. Part of SDXL's micro-conditioning as explained in section 2.2 of [https://huggingface.co/papers/2307.01952] --sdxl-target-size SIZE [SIZE ...], --sdxl-target-sizes SIZE [SIZE ...] One or more Stable Diffusion XL (torch-sdxl) "target-size" micro- conditioning parameters in the format (WIDTH)x(HEIGHT). For most cases, --sdxl-target-size should be set to the desired height and width of the generated image. If not specified it will default to --output-size or the size of any input images. Part of SDXL's micro-conditioning as explained in section 2.2 of [https://huggingface.co/papers/2307.01952] --sdxl-negative-aesthetic-scores FLOAT [FLOAT ...] One or more Stable Diffusion XL (torch-sdxl) "negative-aesthetic-score" micro-conditioning parameters. Part of SDXL's micro-conditioning as explained in section 2.2 of [https://huggingface.co/papers/2307.01952]. Can be used to simulate an aesthetic score of the generated image by influencing the negative text condition. --sdxl-negative-original-sizes SIZE [SIZE ...] One or more Stable Diffusion XL (torch-sdxl) "negative-original-sizes" micro-conditioning parameters. Negatively condition the generation process based on a specific image resolution. Part of SDXL's micro-conditioning as explained in section 2.2 of [https://huggingface.co/papers/2307.01952]. For more information, refer to this issue thread: https://github.com/huggingface/diffusers/issues/4208 --sdxl-negative-target-sizes SIZE [SIZE ...] One or more Stable Diffusion XL (torch-sdxl) "negative-original-sizes" micro-conditioning parameters. To negatively condition the generation process based on a target image resolution. It should be as same as the " --sdxl-target-size" for most cases. Part of SDXL's micro-conditioning as explained in section 2.2 of [https://huggingface.co/papers/2307.01952]. For more information, refer to this issue thread: https://github.com/huggingface/diffusers/issues/4208. --sdxl-negative-crops-coords-top-left COORD [COORD ...] One or more Stable Diffusion XL (torch-sdxl) "negative-crops-coords-top- left" micro-conditioning parameters in the format "0,0". Negatively condition the generation process based on a specific crop coordinates. Part of SDXL's micro-conditioning as explained in section 2.2 of [https://huggingface.co/papers/2307.01952]. For more information, refer to this issue thread: https://github.com/huggingface/diffusers/issues/4208. --sdxl-refiner-prompts PROMPT [PROMPT ...] One or more prompts to try with the SDXL refiner model, by default the refiner model gets the primary prompt, this argument overrides that with a prompt of your choosing. The negative prompt component can be specified with the same syntax as --prompts --sdxl-refiner-clip-skips INTEGER [INTEGER ...] One or more clip skip override values to try for the SDXL refiner, which normally uses the clip skip value for the main model when it is defined by --clip-skips. --sdxl-refiner-second-prompts PROMPT [PROMPT ...] One or more prompts to try with the SDXL refiner models secondary text encoder, by default the refiner model gets the primary prompt passed to its second text encoder, this argument overrides that with a prompt of your choosing. The negative prompt component can be specified with the same syntax as --prompts --sdxl-refiner-aesthetic-scores FLOAT [FLOAT ...] See: --sdxl-aesthetic-scores, applied to SDXL refiner pass. --sdxl-refiner-crops-coords-top-left COORD [COORD ...] See: --sdxl-crops-coords-top-left, applied to SDXL refiner pass. --sdxl-refiner-original-sizes SIZE [SIZE ...] See: --sdxl-refiner-original-sizes, applied to SDXL refiner pass. --sdxl-refiner-target-sizes SIZE [SIZE ...] See: --sdxl-refiner-target-sizes, applied to SDXL refiner pass. --sdxl-refiner-negative-aesthetic-scores FLOAT [FLOAT ...] See: --sdxl-negative-aesthetic-scores, applied to SDXL refiner pass. --sdxl-refiner-negative-original-sizes SIZE [SIZE ...] See: --sdxl-negative-original-sizes, applied to SDXL refiner pass. --sdxl-refiner-negative-target-sizes SIZE [SIZE ...] See: --sdxl-negative-target-sizes, applied to SDXL refiner pass. --sdxl-refiner-negative-crops-coords-top-left COORD [COORD ...] See: --sdxl-negative-crops-coords-top-left, applied to SDXL refiner pass. -hnf FLOAT [FLOAT ...], --sdxl-high-noise-fractions FLOAT [FLOAT ...] One or more high-noise-fraction values for Stable Diffusion XL (torch- sdxl), this fraction of inference steps will be processed by the base model, while the rest will be processed by the refiner model. Multiple values to this argument will result in additional generation steps for each value. In certain situations when the mixture of denoisers algorithm is not supported, such as when using --control-nets and inpainting with SDXL, the inverse proportion of this value IE: (1.0 - high-noise-fraction) becomes the --image-seed-strengths input to the SDXL refiner. (default: [0.8]) -ri INT [INT ...], --sdxl-refiner-inference-steps INT [INT ...] One or more inference steps values for the SDXL refiner when in use. Override the number of inference steps used by the SDXL refiner, which defaults to the value taken from --inference-steps. -rg FLOAT [FLOAT ...], --sdxl-refiner-guidance-scales FLOAT [FLOAT ...] One or more guidance scale values for the SDXL refiner when in use. Override the guidance scale value used by the SDXL refiner, which defaults to the value taken from --guidance-scales. -rgr FLOAT [FLOAT ...], --sdxl-refiner-guidance-rescales FLOAT [FLOAT ...] One or more guidance rescale values for the SDXL refiner when in use. Override the guidance rescale value used by the SDXL refiner, which defaults to the value taken from --guidance-rescales. -sc, --safety-checker Enable safety checker loading, this is off by default. When turned on images with NSFW content detected may result in solid black output. Some pretrained models have no safety checker model present, in that case this option has no effect. -d DEVICE, --device DEVICE cuda / cpu. (default: cuda). Use: cuda:0, cuda:1, cuda:2, etc. to specify a specific GPU. This argument is ignored when using flax, for flax use the environmental variable CUDA_VISIBLE_DEVICES to specify which GPUs are visible to cuda, flax will use every visible GPU. -t DTYPE, --dtype DTYPE Model precision: auto, bfloat16, float16, or float32. (default: auto) -s SIZE, --output-size SIZE Image output size, for txt2img generation, this is the exact output size. The dimensions specified for this value must be aligned by 8 or you will receive an error message. If an --image-seeds URI is used its Seed, Mask, and/or Control component image sources will be resized to this dimension with aspect ratio maintained before being used for generation by default. Unless --no-aspect is specified, width will be fixed and a new height (aligned by 8) will be calculated for the input images. In most cases resizing the image inputs will result in an image output of an equal size to the inputs, except in the case of upscalers and Deep Floyd --model-type values (torch-if*). If only one integer value is provided, that is the value for both dimensions. X/Y dimension values should be separated by "x". This value defaults to 512x512 for Stable Diffusion when no --image- seeds are specified (IE txt2img mode), 1024x1024 for Stable Diffusion XL (SDXL) model types, and 64x64 for --model-type torch-if (Deep Floyd stage 1). Deep Floyd stage 1 images passed to superscaler models (--model-type torch-ifs*) that are specified with the 'floyd' keyword argument in an --image-seeds definition are never resized or processed in any way. -na, --no-aspect This option disables aspect correct resizing of images provided to --image-seeds globally. Seed, Mask, and Control guidance images will be resized to the closest dimension specified by --output-size that is aligned by 8 pixels with no consideration of the source aspect ratio. This can be overriden at the --image-seeds level with the image seed keyword argument 'aspect=true/false'. -o PATH, --output-path PATH Output path for generated images and files. This directory will be created if it does not exist. (default: ./output) -op PREFIX, --output-prefix PREFIX Name prefix for generated images and files. This prefix will be added to the beginning of every generated file, followed by an underscore. -ox, --output-overwrite Enable overwrites of files in the output directory that already exists. The default behavior is not to do this, and instead append a filename suffix: "_duplicate_(number)" when it is detected that the generated file name already exists. -oc, --output-configs Write a configuration text file for every output image or animation. The text file can be used reproduce that particular output image or animation by piping it to dgenerate STDIN or by using the --file option, for example "dgenerate < config.dgen" or "dgenerate --file config.dgen". These files will be written to --output-path and are affected by --output-prefix and --output-overwrite as well. The files will be named after their corresponding image or animation file. Configuration files produced for animation frame images will utilize --frame-start and --frame-end to specify the frame number. -om, --output-metadata Write the information produced by --output-configs to the PNG metadata of each image. Metadata will not be written to animated files (yet). The data is written to a PNG metadata property named DgenerateConfig and can be read using ImageMagick like so: "magick identify -format "%[Property:DgenerateConfig] generated_file.png". -p PROMPT [PROMPT ...], --prompts PROMPT [PROMPT ...] One or more prompts to try, an image group is generated for each prompt, prompt data is split by ; (semi-colon). The first value is the positive text influence, things you want to see. The Second value is negative influence IE. things you don't want to see. Example: --prompts "shrek flying a tesla over detroit; clouds, rain, missiles". (default: [(empty string)]) -cs INTEGER [INTEGER ...], --clip-skips INTEGER [INTEGER ...] One or more clip skip values to try. Clip skip is the number of layers to be skipped from CLIP while computing the prompt embeddings, it must be a value greater than or equal to zero. A value of 1 means that the output of the pre-final layer will be used for computing the prompt embeddings. This is only supported for --model-type values "torch" and "torch-sdxl", including with --control-nets. -se SEED [SEED ...], --seeds SEED [SEED ...] One or more seeds to try, define fixed seeds to achieve deterministic output. This argument may not be used when --gse/--gen-seeds is used. (default: [randint(0, 99999999999999)]) -sei, --seeds-to-images When this option is enabled, each provided --seeds value or value generated by --gen-seeds is used for the corresponding image input given by --image-seeds. If the amount of --seeds given is not identical to that of the amount of --image-seeds given, the seed is determined as: seed = seeds[image_seed_index % len(seeds)], IE: it wraps around. -gse COUNT, --gen-seeds COUNT Auto generate N random seeds to try. This argument may not be used when -se/--seeds is used. -af FORMAT, --animation-format FORMAT Output format when generating an animation from an input video / gif / webp etc. Value must be one of: mp4, png, apng, gif, or webp. You may also specify "frames" to indicate that only frames should be output and no coalesced animation file should be rendered. (default: mp4) -if FORMAT, --image-format FORMAT Output format when writing static images. Any selection other than "png" is not compatible with --output-metadata. Value must be one of: png, apng, blp, bmp, dib, bufr, pcx, dds, ps, eps, gif, grib, h5, hdf, jp2, j2k, jpc, jpf, jpx, j2c, icns, ico, im, jfif, jpe, jpg, jpeg, tif, tiff, mpo, msp, palm, pdf, pbm, pgm, ppm, pnm, pfm, bw, rgb, rgba, sgi, tga, icb, vda, vst, webp, wmf, emf, or xbm. (default: png) -nf, --no-frames Do not write frame images individually when rendering an animation, only write the animation file. This option is incompatible with --animation- format frames. -fs FRAME_NUMBER, --frame-start FRAME_NUMBER Starting frame slice point for animated files (zero-indexed), the specified frame will be included. (default: 0) -fe FRAME_NUMBER, --frame-end FRAME_NUMBER Ending frame slice point for animated files (zero-indexed), the specified frame will be included. -is SEED [SEED ...], --image-seeds SEED [SEED ...] One or more image seed URIs to process, these may consist of URLs or file paths. Videos / GIFs / WEBP files will result in frames being rendered as well as an animated output file being generated if more than one frame is available in the input file. Inpainting for static images can be achieved by specifying a black and white mask image in each image seed string using a semicolon as the separating character, like so: "my-seed-image.png;my- image-mask.png", white areas of the mask indicate where generated content is to be placed in your seed image. Output dimensions specific to the image seed can be specified by placing the dimension at the end of the string following a semicolon like so: "my-seed-image.png;512x512" or "my- seed-image.png;my-image-mask.png;512x512". When using --control-nets, a singular image specification is interpreted as the control guidance image, and you can specify multiple control image sources by separating them with commas in the case where multiple ControlNets are specified, IE: (--image- seeds "control-image1.png, control-image2.png") OR (--image-seeds "seed.png;control=control-image1.png, control-image2.png"). Using --control-nets with img2img or inpainting can be accomplished with the syntax: "my-seed-image.png;mask=my-image-mask.png;control=my-control- image.png;resize=512x512". The "mask" and "resize" arguments are optional when using --control-nets. Videos, GIFs, and WEBP are also supported as inputs when using --control-nets, even for the "control" argument. --image-seeds is capable of reading from multiple animated files at once or any combination of animated files and images, the animated file with the least amount of frames dictates how many frames are generated and static images are duplicated over the total amount of frames. The keyword argument "aspect" can be used to determine resizing behavior when the global argument --output-size or the local keyword argument "resize" is specified, it is a boolean argument indicating whether aspect ratio of the input image should be respected or ignored. The keyword argument "floyd" can be used to specify images from a previous deep floyd stage when using --model-type torch-ifs*. When keyword arguments are present, all applicable images such as "mask", "control", etc. must also be defined with keyword arguments instead of with the short syntax. -sip PROCESSOR_URI [PROCESSOR_URI ...], --seed-image-processors PROCESSOR_URI [PROCESSOR_URI ...] Specify one or more image processor actions to preform on the primary image specified by --image-seeds. For example: --seed-image-processors "flip" "mirror" "grayscale". To obtain more information about what image processors are available and how to use them, see: --image-processor-help. -mip PROCESSOR_URI [PROCESSOR_URI ...], --mask-image-processors PROCESSOR_URI [PROCESSOR_URI ...] Specify one or more image processor actions to preform on the inpaint mask image specified by --image-seeds. For example: --mask-image-processors "invert". To obtain more information about what image processors are available and how to use them, see: --image-processor-help. -cip PROCESSOR_URI [PROCESSOR_URI ...], --control-image-processors PROCESSOR_URI [PROCESSOR_URI ...] Specify one or more image processor actions to preform on the control image specified by --image-seeds, this option is meant to be used with --control-nets. Example: --control-image-processors "canny;lower=50;upper=100". The delimiter "+" can be used to specify a different processor group for each image when using multiple control images with --control-nets. For example if you have --image-seeds "img1.png, img2.png" or --image-seeds "...;control=img1.png, img2.png" specified and multiple ControlNet models specified with --control-nets, you can specify processors for those control images with the syntax: (--control-image-processors "processes-img1" + "processes-img2"), this syntax also supports chaining of processors, for example: (--control- image-processors "first-process-img1" "second-process-img1" + "process- img2"). The amount of specified processors must not exceed the amount of specified control images, or you will receive a syntax error message. Images which do not have a processor defined for them will not be processed, and the plus character can be used to indicate an image is not to be processed and instead skipped over when that image is a leading element, for example (--control-image-processors + "process-second") would indicate that the first control guidance image is not to be processed, only the second. To obtain more information about what image processors are available and how to use them, see: --image-processor-help. --image-processor-help [PROCESSOR_NAME ...] Use this option alone (or with --plugin-modules) and no model specification in order to list available image processor module names. Specifying one or more module names after this option will cause usage documentation for the specified modules to be printed. -pp PROCESSOR_URI [PROCESSOR_URI ...], --post-processors PROCESSOR_URI [PROCESSOR_URI ...] Specify one or more image processor actions to preform on generated output before it is saved. For example: --post-processors "upcaler;model=4x_ESRGAN.pth". To obtain more information about what processors are available and how to use them, see: --image-processor-help. -iss FLOAT [FLOAT ...], --image-seed-strengths FLOAT [FLOAT ...] One or more image strength values to try when using --image-seeds for img2img or inpaint mode. Closer to 0 means high usage of the seed image (less noise convolution), 1 effectively means no usage (high noise convolution). Low values will produce something closer or more relevant to the input image, high values will give the AI more creative freedom. (default: [0.8]) -uns INTEGER [INTEGER ...], --upscaler-noise-levels INTEGER [INTEGER ...] One or more upscaler noise level values to try when using the super resolution upscaler --model-type torch-upscaler-x4 or torch-ifs. Specifying this option for --model-type torch-upscaler-x2 will produce an error message. The higher this value the more noise is added to the image before upscaling (similar to --image-seed-strengths). (default: [20 for x4, 250 for torch-ifs/torch-ifs-img2img, 0 for torch-ifs inpainting mode]) -gs FLOAT [FLOAT ...], --guidance-scales FLOAT [FLOAT ...] One or more guidance scale values to try. Guidance scale effects how much your text prompt is considered. Low values draw more data from images unrelated to text prompt. (default: [5]) -igs FLOAT [FLOAT ...], --image-guidance-scales FLOAT [FLOAT ...] One or more image guidance scale values to try. This can push the generated image towards the initial image when using --model-type *-pix2pix models, it is unsupported for other model types. Use in conjunction with --image-seeds, inpainting (masks) and --control-nets are not supported. Image guidance scale is enabled by setting image-guidance- scale > 1. Higher image guidance scale encourages generated images that are closely linked to the source image, usually at the expense of lower image quality. Requires a value of at least 1. (default: [1.5]) -gr FLOAT [FLOAT ...], --guidance-rescales FLOAT [FLOAT ...] One or more guidance rescale factors to try. Proposed by [Common Diffusion Noise Schedules and Sample Steps are Flawed](https://arxiv.org/pdf/2305.08891.pdf) "guidance_scale" is defined as "φ" in equation 16. of [Common Diffusion Noise Schedules and Sample Steps are Flawed] (https://arxiv.org/pdf/2305.08891.pdf). Guidance rescale factor should fix overexposure when using zero terminal SNR. This is supported for basic text to image generation when using --model-type "torch" but not inpainting, img2img, or --control-nets. When using --model-type "torch-sdxl" it is supported for basic generation, inpainting, and img2img, unless --control-nets is specified in which case only inpainting is supported. It is supported for --model-type "torch- sdxl-pix2pix" but not --model-type "torch-pix2pix". (default: [0.0]) -ifs INTEGER [INTEGER ...], --inference-steps INTEGER [INTEGER ...] One or more inference steps values to try. The amount of inference (de- noising) steps effects image clarity to a degree, higher values bring the image closer to what the AI is targeting for the content of the image. Values between 30-40 produce good results, higher values may improve image quality and or change image content. (default: [30]) -mc EXPR [EXPR ...], --cache-memory-constraints EXPR [EXPR ...] Cache constraint expressions describing when to clear all model caches automatically (DiffusionPipeline, VAE, and ControlNet) considering current memory usage. If any of these constraint expressions are met all models cached in memory will be cleared. Example, and default value: "used_percent > 70" For Syntax See: [https://dgenerate.readthedocs.io/en/v 3.7.1/dgenerate_submodules.html#dgenerate.pipelinewrapper.CACHE_MEMORY_CON STRAINTS] -pmc EXPR [EXPR ...], --pipeline-cache-memory-constraints EXPR [EXPR ...] Cache constraint expressions describing when to automatically clear the in memory DiffusionPipeline cache considering current memory usage, and estimated memory usage of new models that are about to enter memory. If any of these constraint expressions are met all DiffusionPipeline objects cached in memory will be cleared. Example, and default value: "pipeline_size > (available * 0.75)" For Syntax See: [https://dgenerate.re adthedocs.io/en/v3.7.1/dgenerate_submodules.html#dgenerate.pipelinewrapper .PIPELINE_CACHE_MEMORY_CONSTRAINTS] -umc EXPR [EXPR ...], --unet-cache-memory-constraints EXPR [EXPR ...] Cache constraint expressions describing when to automatically clear the in memory UNet cache considering current memory usage, and estimated memory usage of new UNet models that are about to enter memory. If any of these constraint expressions are met all UNet models cached in memory will be cleared. Example, and default value: "unet_size > (available * 0.75)" For Syntax See: [https://dgenerate.readthedocs.io/en/v3.7.1/dgenerate_submodul es.html#dgenerate.pipelinewrapper.UNET_CACHE_MEMORY_CONSTRAINTS] -vmc EXPR [EXPR ...], --vae-cache-memory-constraints EXPR [EXPR ...] Cache constraint expressions describing when to automatically clear the in memory VAE cache considering current memory usage, and estimated memory usage of new VAE models that are about to enter memory. If any of these constraint expressions are met all VAE models cached in memory will be cleared. Example, and default value: "vae_size > (available * 0.75)" For Syntax See: [https://dgenerate.readthedocs.io/en/v3.7.1/dgenerate_submodul es.html#dgenerate.pipelinewrapper.VAE_CACHE_MEMORY_CONSTRAINTS] -cmc EXPR [EXPR ...], --control-net-cache-memory-constraints EXPR [EXPR ...] Cache constraint expressions describing when to automatically clear the in memory ControlNet cache considering current memory usage, and estimated memory usage of new ControlNet models that are about to enter memory. If any of these constraint expressions are met all ControlNet models cached in memory will be cleared. Example, and default value: "control_net_size > (available * 0.75)" For Syntax See: [https://dgenerate.readthedocs.io/en/v 3.7.1/dgenerate_submodules.html#dgenerate.pipelinewrapper.CONTROL_NET_CACH E_MEMORY_CONSTRAINTS]

pip install pipxpipx ensurepath# Log out and log back in so PATH takes effect

pipx install dgenerate ^--pip-args "--extra-index-url https://download.pytorch.org/whl/cu121/"# If you want a specific versionpipx install dgenerate==3.7.1 ^--pip-args "--extra-index-url https://download.pytorch.org/whl/cu121/"# You can install without pipx into your own environment like sopip install dgenerate==3.7.1 --extra-index-url https://download.pytorch.org/whl/cu121/

# in the top of the repo make# an environment and activate itpython -m venv venvvenv\Scripts\activate# Install with pip into the environmentpip install --editable .[dev] --extra-index-url https://download.pytorch.org/whl/cu121/

# Images are output to the "output" folder# in the current working directory by defaultdgenerate --helpdgenerate stabilityai/stable-diffusion-2-1 ^--prompts "an astronaut riding a horse" ^--output-path output ^--inference-steps 40 ^--guidance-scales 10

sudo apt install python3.10 python3-pip pipx python3.10-venv python3-wheelpipx ensurepathsource ~/.bashrc

pipx install dgenerate \--pip-args "--extra-index-url https://download.pytorch.org/whl/cu121/"# With flax/jax supportpipx install dgenerate[flax] \--pip-args "--extra-index-url https://download.pytorch.org/whl/cu121/ \-f https://storage.googleapis.com/jax-releases/jax_cuda_releases.html"# If you want a specific versionpipx install dgenerate==3.7.1 \--pip-args "--extra-index-url https://download.pytorch.org/whl/cu121/"# Specific version with flax/jax supportpipx install dgenerate[flax]==3.7.1 \--pip-args "--extra-index-url https://download.pytorch.org/whl/cu121/ \-f https://storage.googleapis.com/jax-releases/jax_cuda_releases.html"# You can install without pipx into your own environment like sopip3 install dgenerate==3.7.1 --extra-index-url https://download.pytorch.org/whl/cu121/# Or with flaxpip3 install dgenerate[flax]==3.7.1 --extra-index-url https://download.pytorch.org/whl/cu121/ \-f https://storage.googleapis.com/jax-releases/jax_cuda_releases.html

# in the top of the repo make# an environment and activate itpython3 -m venv venvsource venv/bin/activate# Install with pip into the environmentpip3 install --editable .[dev] --extra-index-url https://download.pytorch.org/whl/cu121/# With flax if you wantpip3 install --editable .[dev,flax] --extra-index-url https://download.pytorch.org/whl/cu121/ \-f https://storage.googleapis.com/jax-releases/jax_cuda_releases.html

# Images are output to the "output" folder# in the current working directory by defaultdgenerate --helpdgenerate stabilityai/stable-diffusion-2-1 \--prompts "an astronaut riding a horse" \--output-path output \--inference-steps 40 \--guidance-scales 10

# these all expand from your system environment# if the variable is not set, they expand to nothing\print $VARIABLE\print ${VARIABLE}\print %VARIABLE%

#! /usr/bin/env dgenerate --file#! dgenerate 3.7.1# If a hash-bang version is provided in the format above# a warning will be produced if the version you are running# is not compatible (SemVer), this can be used anywhere in the# config file, a line number will be mentioned in the warning when the# version check fails# Comments in the file will be ignored# Each dgenerate invocation in the config begins with the path to a model,# IE. the first argument when using dgenerate from the command line, the# rest of the options that follow are the options to dgenerate that you# would use on the command line# Guarantee unique file names are generated under the output directory by specifying unique seedsstabilityai/stable-diffusion-2-1 --prompts "an astronaut riding a horse" --seeds 41509644783027 --output-path output --inference-steps 30 --guidance-scales 10stabilityai/stable-diffusion-2-1 --prompts "a cowboy riding a horse" --seeds 78553317097366 --output-path output --inference-steps 30 --guidance-scales 10stabilityai/stable-diffusion-2-1 --prompts "a martian riding a horse" --seeds 22797399276707 --output-path output --inference-steps 30 --guidance-scales 10# Guarantee that no file name collisions happen by specifying different output paths for each invocationstabilityai/stable-diffusion-2-1 --prompts "an astronaut riding a horse" --output-path unique_output_1 --inference-steps 30 --guidance-scales 10stabilityai/stable-diffusion-2-1 --prompts "a cowboy riding a horse" --output-path unique_output_2 --inference-steps 30 --guidance-scales 10# Multiline continuations are possible implicitly for argument# switches IE lines starting with '-'stabilityai/stable-diffusion-2-1 --prompts "a martian riding a horse"--output-path unique_output_3 # there can be comments at the end of lines--inference-steps 30 \ # this comment is also ignored# There can be comments or newlines within the continuation# but you must provide \ on the previous line to indicate that# it is going to happen--guidance-scales 10# The continuation ends (on the next line) when the last line does# not end in \ or start with -# the ability to use tail comments means that escaping of the # is sometimes# necessary when you want it to appear literally, see: examples/config_syntax/tail-comments-config.dgen# for examples.# Configuration directives provide extra functionality in a config, a directive# invocation always starts with a backslash# A clear model cache directive can be used inbetween invocations if cached models that# are no longer needed in your generation pipeline start causing out of memory issues\clear_model_cache# Additionally these other directives exist to clear user loaded models# out of dgenerates in memory cache individually# Clear specifically diffusion pipelines\clear_pipeline_cache# Clear specifically user specified UNet models\clear_unet_cache# Clear specifically user specified VAE models\clear_vae_cache# Clear specifically ControlNet models\clear_control_net_cache# This model was used before but will have to be fully instantiated from scratch again# after a cache flush which may take some timestabilityai/stable-diffusion-2-1 --prompts "a martian riding a horse"--output-path unique_output_4

#! /usr/bin/env dgenerate --file#! dgenerate 3.7.1# Invocation will proceed as normalstabilityai/stable-diffusion-2-1 --prompts "a man walking on the moon without a space suit"# Print all set template variables\templates_help

Config template variables are: Name: "last_model_path" Type: typing.Optional[str] Value: stabilityai/stable-diffusion-2-1 Name: "last_subfolder" Type: typing.Optional[str] Value: None Name: "last_sdxl_refiner_uri" Type: typing.Optional[str] Value: None Name: "last_sdxl_refiner_edit" Type: typing.Optional[bool] Value: None Name: "last_batch_size" Type: typing.Optional[int] Value: 1 Name: "last_batch_grid_size" Type: typing.Optional[tuple[int, int]] Value: None Name: "last_prompts" Type: collections.abc.Sequence[dgenerate.prompt.Prompt] Value: ['a man walking on the moon without a space suit'] Name: "last_sdxl_second_prompts" Type: typing.Optional[collections.abc.Sequence[dgenerate.prompt.Prompt]] Value: [] Name: "last_sdxl_refiner_prompts" Type: typing.Optional[collections.abc.Sequence[dgenerate.prompt.Prompt]] Value: [] Name: "last_sdxl_refiner_second_prompts" Type: typing.Optional[collections.abc.Sequence[dgenerate.prompt.Prompt]] Value: [] Name: "last_seeds" Type: collections.abc.Sequence[int] Value: [98030306583037] Name: "last_seeds_to_images" Type: <class 'bool'> Value: False Name: "last_guidance_scales" Type: collections.abc.Sequence[float] Value: [5] Name: "last_inference_steps" Type: collections.abc.Sequence[int] Value: [30] Name: "last_clip_skips" Type: typing.Optional[collections.abc.Sequence[int]] Value: [] Name: "last_sdxl_refiner_clip_skips" Type: typing.Optional[collections.abc.Sequence[int]] Value: [] Name: "last_image_seeds" Type: typing.Optional[collections.abc.Sequence[str]] Value: [] Name: "last_parsed_image_seeds" Type: typing.Optional[collections.abc.Sequence[dgenerate.mediainput.ImageSeedParseResult]] Value: [] Name: "last_image_seed_strengths" Type: typing.Optional[collections.abc.Sequence[float]] Value: [] Name: "last_upscaler_noise_levels" Type: typing.Optional[collections.abc.Sequence[int]] Value: [] Name: "last_guidance_rescales" Type: typing.Optional[collections.abc.Sequence[float]] Value: [] Name: "last_image_guidance_scales" Type: typing.Optional[collections.abc.Sequence[float]] Value: [] Name: "last_s_cascade_decoder_uri" Type: typing.Optional[str] Value: None Name: "last_s_cascade_decoder_prompts" Type: typing.Optional[collections.abc.Sequence[dgenerate.prompt.Prompt]] Value: [] Name: "last_s_cascade_decoder_inference_steps" Type: typing.Optional[collections.abc.Sequence[int]] Value: [] Name: "last_s_cascade_decoder_guidance_scales" Type: typing.Optional[collections.abc.Sequence[float]] Value: [] Name: "last_sdxl_high_noise_fractions" Type: typing.Optional[collections.abc.Sequence[float]] Value: [] Name: "last_sdxl_refiner_inference_steps" Type: typing.Optional[collections.abc.Sequence[int]] Value: [] Name: "last_sdxl_refiner_guidance_scales" Type: typing.Optional[collections.abc.Sequence[float]] Value: [] Name: "last_sdxl_refiner_guidance_rescales" Type: typing.Optional[collections.abc.Sequence[float]] Value: [] Name: "last_sdxl_aesthetic_scores" Type: typing.Optional[collections.abc.Sequence[float]] Value: [] Name: "last_sdxl_original_sizes" Type: typing.Optional[collections.abc.Sequence[tuple[int, int]]] Value: [] Name: "last_sdxl_target_sizes" Type: typing.Optional[collections.abc.Sequence[tuple[int, int]]] Value: [] Name: "last_sdxl_crops_coords_top_left" Type: typing.Optional[collections.abc.Sequence[tuple[int, int]]] Value: [] Name: "last_sdxl_negative_aesthetic_scores" Type: typing.Optional[collections.abc.Sequence[float]] Value: [] Name: "last_sdxl_negative_original_sizes" Type: typing.Optional[collections.abc.Sequence[tuple[int, int]]] Value: [] Name: "last_sdxl_negative_target_sizes" Type: typing.Optional[collections.abc.Sequence[tuple[int, int]]] Value: [] Name: "last_sdxl_negative_crops_coords_top_left" Type: typing.Optional[collections.abc.Sequence[tuple[int, int]]] Value: [] Name: "last_sdxl_refiner_aesthetic_scores" Type: typing.Optional[collections.abc.Sequence[float]] Value: [] Name: "last_sdxl_refiner_original_sizes" Type: typing.Optional[collections.abc.Sequence[tuple[int, int]]] Value: [] Name: "last_sdxl_refiner_target_sizes" Type: typing.Optional[collections.abc.Sequence[tuple[int, int]]] Value: [] Name: "last_sdxl_refiner_crops_coords_top_left" Type: typing.Optional[collections.abc.Sequence[tuple[int, int]]] Value: [] Name: "last_sdxl_refiner_negative_aesthetic_scores" Type: typing.Optional[collections.abc.Sequence[float]] Value: [] Name: "last_sdxl_refiner_negative_original_sizes" Type: typing.Optional[collections.abc.Sequence[tuple[int, int]]] Value: [] Name: "last_sdxl_refiner_negative_target_sizes" Type: typing.Optional[collections.abc.Sequence[tuple[int, int]]] Value: [] Name: "last_sdxl_refiner_negative_crops_coords_top_left" Type: typing.Optional[collections.abc.Sequence[tuple[int, int]]] Value: [] Name: "last_unet_uri" Type: typing.Optional[str] Value: None Name: "last_second_unet_uri" Type: typing.Optional[str] Value: None Name: "last_vae_uri" Type: typing.Optional[str] Value: None Name: "last_vae_tiling" Type: <class 'bool'> Value: False Name: "last_vae_slicing" Type: <class 'bool'> Value: False Name: "last_lora_uris" Type: typing.Optional[collections.abc.Sequence[str]] Value: [] Name: "last_textual_inversion_uris" Type: typing.Optional[collections.abc.Sequence[str]] Value: [] Name: "last_control_net_uris" Type: typing.Optional[collections.abc.Sequence[str]] Value: [] Name: "last_scheduler" Type: typing.Optional[str] Value: None Name: "last_sdxl_refiner_scheduler" Type: typing.Optional[str] Value: None Name: "last_s_cascade_decoder_scheduler" Type: typing.Optional[str] Value: None Name: "last_safety_checker" Type: <class 'bool'> Value: False Name: "last_model_type" Type: <enum 'ModelType'> Value: ModelType.TORCH Name: "last_device" Type: <class 'str'> Value: cuda Name: "last_dtype" Type: <enum 'DataType'> Value: DataType.AUTO Name: "last_revision" Type: <class 'str'> Value: main Name: "last_variant" Type: typing.Optional[str] Value: None Name: "last_output_size" Type: typing.Optional[tuple[int, int]] Value: (512, 512) Name: "last_no_aspect" Type: <class 'bool'> Value: False Name: "last_output_path" Type: <class 'str'> Value: output Name: "last_output_prefix" Type: typing.Optional[str] Value: None Name: "last_output_overwrite" Type: <class 'bool'> Value: False Name: "last_output_configs" Type: <class 'bool'> Value: False Name: "last_output_metadata" Type: <class 'bool'> Value: False Name: "last_animation_format" Type: <class 'str'> Value: mp4 Name: "last_image_format" Type: <class 'str'> Value: png Name: "last_no_frames" Type: <class 'bool'> Value: False Name: "last_frame_start" Type: <class 'int'> Value: 0 Name: "last_frame_end" Type: typing.Optional[int] Value: None Name: "last_auth_token" Type: typing.Optional[str] Value: None Name: "last_seed_image_processors" Type: typing.Optional[collections.abc.Sequence[str]] Value: [] Name: "last_mask_image_processors" Type: typing.Optional[collections.abc.Sequence[str]] Value: [] Name: "last_control_image_processors" Type: typing.Optional[collections.abc.Sequence[str]] Value: [] Name: "last_post_processors" Type: typing.Optional[collections.abc.Sequence[str]] Value: [] Name: "last_offline_mode" Type: <class 'bool'> Value: False Name: "last_model_cpu_offload" Type: <class 'bool'> Value: False Name: "last_model_sequential_offload" Type: <class 'bool'> Value: False Name: "last_sdxl_refiner_cpu_offload" Type: typing.Optional[bool] Value: None Name: "last_sdxl_refiner_sequential_offload" Type: typing.Optional[bool] Value: None Name: "last_s_cascade_decoder_cpu_offload" Type: typing.Optional[bool] Value: None Name: "last_s_cascade_decoder_sequential_offload" Type: typing.Optional[bool] Value: None Name: "last_plugin_module_paths" Type: collections.abc.Sequence[str] Value: [] Name: "last_verbose" Type: <class 'bool'> Value: False Name: "last_cache_memory_constraints" Type: typing.Optional[collections.abc.Sequence[str]] Value: [] Name: "last_pipeline_cache_memory_constraints" Type: typing.Optional[collections.abc.Sequence[str]] Value: [] Name: "last_unet_cache_memory_constraints" Type: typing.Optional[collections.abc.Sequence[str]] Value: [] Name: "last_vae_cache_memory_constraints" Type: typing.Optional[collections.abc.Sequence[str]] Value: [] Name: "last_control_net_cache_memory_constraints" Type: typing.Optional[collections.abc.Sequence[str]] Value: [] Name: "last_images" Type: collections.abc.Iterable[str] Value: <dgenerate.renderloop.RenderLoop.written_images.<locals>.Iterable object at ...> Name: "last_animations" Type: collections.abc.Iterable[str] Value: <dgenerate.renderloop.RenderLoop.written_animations.<locals>.Iterable object at ...> Name: "injected_args" Type: collections.abc.Sequence[str] Value: [] Name: "injected_device" Type: typing.Optional[str] Value: None Name: "injected_verbose" Type: typing.Optional[bool] Value: False Name: "injected_plugin_modules" Type: typing.Optional[collections.abc.Sequence[str]] Value: [] Name: "saved_modules" Type: dict[str, dict[str, typing.Any]] Value: {} Name: "glob" Type: <class 'module'> Value: <module 'glob'> Name: "path" Type: <class 'module'> Value: <module 'ntpath' (frozen)>

Available config template functions: abs(args, kwargs) all(args, kwargs) any(args, kwargs) ascii(args, kwargs) bin(args, kwargs) bool(args, kwargs) bytearray(args, kwargs) bytes(args, kwargs) callable(args, kwargs) chr(args, kwargs) complex(args, kwargs) cwd() -> str dict(args, kwargs) divmod(args, kwargs) download(url: str, output: str | None = None, overwrite: bool = False, text: bool = False) -> str enumerate(args, kwargs) filter(args, kwargs) first(iterable: collections.abc.Iterable[typing.Any]) -> typing.Any float(args, kwargs) format(args, kwargs) format_prompt(prompts: dgenerate.prompt.Prompt | collections.abc.Iterable[dgenerate.prompt.Prompt]) -> str format_size(size: collections.abc.Iterable[int]) -> str frozenset(args, kwargs) gen_seeds(n: int) -> list[str] getattr(args, kwargs) hasattr(args, kwargs) hash(args, kwargs) hex(args, kwargs) int(args, kwargs) iter(args, kwargs) last(iterable: list | collections.abc.Iterable[typing.Any]) -> typing.Any len(args, kwargs) list(args, kwargs) map(args, kwargs) max(args, kwargs) min(args, kwargs) next(args, kwargs) object(args, kwargs) oct(args, kwargs) ord(args, kwargs) pow(args, kwargs) quote(strings: str | collections.abc.Iterable[typing.Any]) -> str range(args, kwargs) repr(args, kwargs) reversed(args, kwargs) round(args, kwargs) set(args, kwargs) slice(args, kwargs) sorted(args, kwargs) str(args, kwargs) sum(args, kwargs) tuple(args, kwargs) type(args, kwargs) unquote(strings: str | collections.abc.Iterable[typing.Any], expand: bool = False) -> list zip(args, kwargs)

Available config directives: "\help" "\templates_help" "\directives_help" "\functions_help" "\image_processor_help" "\clear_model_cache" "\clear_pipeline_cache" "\clear_unet_cache" "\clear_vae_cache" "\clear_control_net_cache" "\save_modules" "\use_modules" "\clear_modules" "\gen_seeds" "\download" "\pwd" "\ls" "\cd" "\pushd" "\popd" "\exec" "\mv" "\cp" "\mkdir" "\rmdir" "\rm" "\exit" "\image_process" "\import_plugins" "\set" "\sete" "\setp" "\unset" "\print" "\echo"

#! /usr/bin/env dgenerate --file#! dgenerate 3.7.1# You can define your own template variables with the \set directive# the \set directive does not do any shell args parsing on its value# operand, meaning the quotes will be in the string that is assigned# to the variable my_prompt\set my_prompt "an astronaut riding a horse; bad quality"# If your variable is long you can use continuation, note that# continuation replaces newlines and surrounding whitespace# with a single space\set my_prompt "my very very very very very very very \ very very very very very very very very \ long long long long long prompt"# You can print to the console with templating using the \print directive# for debugging purposes\print {{ my_prompt }}# The \setp directive can be used to define python literal template variables\setp my_list [1, 2, 3, 4]\print {{ my_list | join(' ') }}# Literals defined by \setp can reference other template variables by name.# the following creates a nested list\setp my_list [1, 2, my_list, 4]\print {{ my_list }}# \setp can evaluate template functions\setp directory_content glob.glob('*')\setp current_directory cwd()# the \gen_seeds directive can be used to store a list of# random seed integers into a template variable.# (they are strings for convenience)\gen_seeds my_seeds 10\print {{ my_seeds | join(' ') }}# An invocation sets various template variables related to its# execution once it is finished runningstabilityai/stable-diffusion-2-1 --prompts {{ my_prompt }} --gen-seeds 5# Print a quoted filename of the last image produced by the last invocation# This could potentially be passed to --image-seeds of the next invocation# If you wanted to run another pass over the last image that was produced\print {{ quote(last(last_images)) }}# you can also get the first image easily with the function "first"\print {{ quote(first(last_images)) }}# if you want to append a mask image file name\print "{{ last(last_images) }};my-mask.png"# Print a list of properly quoted filenames produced by the last# invocation separated by spaces if there is multiple, this could# also be passed to --image-seeds# in the case that you have generated an animated output with frame# output enabled, this will contain paths to the frames\print {{ quote(last_images) }}# For loops are possible\print {% for image in last_images %}{{ quote(image) }} {% endfor %}# For loops are possible with normal continuation# when not using a heredoc template continuation (mentioned below),# such as when the loop occurs in the body of a directive or a# dgenerate invocation, however this sort of continuation usage will# replace newlines and whitespace with a single space.# IE this template will be: "{% for image in last_images %} {{ quote(image) }} {% endfor %}"\print {% for image in last_images %} \ {{ quote(image) }} \ {% endfor %}# Access to the last prompt is available in a parsed representation# via "last_prompt", which can be formatted properly for reuse# by using the function "format_prompt"stabilityai/stable-diffusion-2-1 --prompts {{ format_prompt(last(last_prompts)) }}# You can get only the positive or negative part if you want via the "positive"# and "negative" properties on a prompt object, these attributes are not# quoted so you need to quote them one way or another, preferably using the# dgenerate template function "quote" which will shell quote any special# characters that the argument parser is not going to understandstabilityai/stable-diffusion-2-1 --prompts {{ quote(last(last_prompts).positive) }}# "last_prompts" returns all the prompts used in the last invocation as a list# the "format_prompt" function can also work on a liststabilityai/stable-diffusion-2-1 --prompts "prompt 1" "prompt 2" "prompt 3"stabilityai/stable-diffusion-2-1 --prompts {{ format_prompt(last_prompts) }}# Execute additional config with full templating.# The sequence !END is interpreted as the end of a# template continuation, a template continuation is# started when a line begins with the character {# and is effectively a heredoc, in that all whitespace# within is preserved including newlines{% for image in last_images %} stabilityai/stable-diffusion-2-1 --image-seeds {{ quote(image) }} --prompts {{ my_prompt }}{% endfor %} !END# Multiple lines can be used with a template continuation# the inside of the template will be expanded to raw config# and then be ran, so make sure to use line continuations within# where they are necessary as you would do in the top level of# a config file. The whole of the template continuation is# processed by Jinja, from { to !END, so only one !END is# ever necessary after the external template# when dealing with nested templates{% for image in last_images %} stabilityai/stable-diffusion-2-1 --image-seeds {{ quote(image) }} --prompts {{ my_prompt }}{% endfor %} !END# The above are both basically equivalent to thisstabilityai/stable-diffusion-2-1 --image-seeds {{ quote(last_images) }} --prompts {{ my_prompt }}# You can save modules from the main pipeline used in the last invocation# for later reuse using the \save_modules directive, the first argument# is a variable name and the rest of the arguments are diffusers pipeline# module names to save to the variable name, this is an advanced usage# and requires some understanding of the diffusers library to utilize correctlystabilityai/stable-diffusion-2-1--variant fp16--dtype float16--prompts "an astronaut walking on the moon"--safety-checker--output-size 512\save_modules stage_1_modules feature_extractor safety_checker# that saves the feature_extractor module object in the pipeline above,# you can specify multiple module names to save if desired# Possible Module Names:# unet# vae# text_encoder# text_encoder_2# tokenizer# tokenizer_2# safety_checker# feature_extractor# controlnet# scheduler# To use the saved modules in the next invocation use \use_modules\use_modules stage_1_modules# now the next invocation will use those modules instead of loading them from internal# in memory cache, disk, or huggingfacestabilityai/stable-diffusion-x4-upscaler--variant fp16--dtype float16--model-type torch-upscaler-x4--prompts {{ format_prompt(last_prompts) }}--image-seeds {{ quote(last_images) }}--vae-tiling# you should clear out the saved modules if you no longer need them# and your config file is going to continue, or if the dgenerate# process is going to be kept alive for some reason such as in# some library usage scenarios, or perhaps if you are using it# like a server that reads from stdin :)\clear_modules stage_1_modules

#! /usr/bin/env dgenerate --file#! dgenerate 3.7.1\set my_variable "I am an incomplete string and this is completely fine because I am a raw value# prints exactly what is above, including the quote at the beginning\print {{ my_variable }}# add a quote to the end of the string using templates\set my_variable {{ my_variable }}"# prints a fully quoted string\print {{ my_variable }}

#! /usr/bin/env dgenerate --file#! dgenerate 3.7.1# lets pretend the directory "my_files" is full of files\sete my_variable --argument my_files/*# prints the python array ['--argument', 'my_files/file1', 'my_files/file2', ...]\print {{ my_variable }}# Templates and environmental variable references# are also parsed in the \sete directive, just as they are with \set\set directory my_files\sete my_variable --argument {{ directory }}/*

#! /usr/bin/env dgenerate --file#! dgenerate 3.7.1\setp my_variable 10*10# prints 100\print {{ my_variable }}# you can reference variables defined in the environment\setp my_variable [my_variable, my_variable*2]# prints [100, 200]\print {{ my_variable }}# all forms of python comprehensions are supported# such as list, dict, and set comprehensions\setp my_variable [i for i in range(0,5)]# prints [0, 1, 2, 3, 4]\print {{ my_variable }}# declare a literal string value\setp my_variable "my string value"# prints the string without quotes included, the string was parsed\print {{ my_variable }}# templates and environmental variable references# are also expanded in \setp values\setp my_variable [my_variable, "{{ my_variable }}"]# prints ["my string value", "my string value"]\print {{ my_variable }}# my_variable is a literal list so it can be# looped over with a jinja template continuation{% for value in my_variable %} \print {{ value }}{% endfor %} !END

#! /usr/bin/env dgenerate --file#! dgenerate 3.7.1# globbing can be preformed via shell expansion or using# the glob module inside jinja templates# note that shell globbing and home directory expansion# does not occur inside quoted strings# \echo can be use to show the results of globbing that# occurs during shell expansion. \print does not preform shell# expansion nor does \set or \setp, all other directives do, as well# as dgenerate invocations# shell globs which produce 0 files are considered an error\echo ../media/*.png\echo ~# \sete can be used to set a template variable to the result# of one or more shell globs\sete myfiles ../media/*.png# with Jinja2:# The most basic usage is full expansion of every file\set myfiles {{ quote(glob.glob('../media/*.png')) }}\print {{ myfiles }}# If you have a LOT of files, you may want to# process them using an iterator like so{% for file in glob.iglob('../media/*.png') %} \print {{ quote(file) }}{% endfor %} !END# usage of os.path via path\print {{ path.abspath('.') }}# Simple inline usagestabilityai/stable-diffusion-2-1--variant fp16--dtype float16--prompts "In the style of picaso"--image-seeds {{ quote(glob.glob('../media/*.png')) }}--output-path {{ quote(path.join(path.abspath('.'), 'output')) }}# equivalentstabilityai/stable-diffusion-2-1--variant fp16--dtype float16--prompts "In the style of picaso"--image-seeds ../media/*.png--output-path ./output

#! /usr/bin/env dgenerate --file#! dgenerate 3.7.1# the text after \print(space) will be printed verbatim\print I am a raw value, I have no ability to * glob# Print the PATH environmental variable\set header Path Elements:\print {{ header }} $PATH\print {{ header }} ${PATH}\print {{ header }} %PATH%

#! /usr/bin/env dgenerate --file#! dgenerate 3.7.1# lets pretend "directory" is full of files# this prints: directory/file1 directory/file2 ...\echo directory/*# Templates and environmental variables are expanded# this prints: Files: directory/file1 directory/file2 ...\set header Files:\echo {{ header }} directory/*

#! /usr/bin/env dgenerate --file#! dgenerate 3.7.1# print the help message of --sub-command image-process, this does# not cause the config to exit\image_process --help\set myfiles {{ quote(glob.glob('my_images/*.png')) }}# this will create the directory "upscaled"# the files will be named "upscaled/FILENAME_processed_1.png" "upscaled/FILENAME_processed_2.png" ...\image_process {{ myfiles }} \--output upscaled/--processors upscaler;model=https://github.com/xinntao/Real-ESRGAN/releases/download/v0.2.5.0/realesr-general-x4v3.pth# the last_images template variable will be set, last_animations is also usable if# animations were written. In the case that you have generated an animated output with frame# output enabled, this will contain paths to the frames\print {{ quote(last_images) }}

#! /usr/bin/env dgenerate --file#! dgenerate 3.7.1# run dgenerate as a subprocess, read a config# and send stdout and stderr to a file\exec dgenerate < my_config.dgen &> log.txt# chaining processes together with pipes is supported# this example emulates 'cat' on Windows using cmd\exec cmd /c "type my_config.dgen" | dgenerate &> log.txt# on a Unix platform you could simply use cat\exec cat my_config.dgen | dgenerate &> log.txt

#! /usr/bin/env dgenerate --file#! dgenerate 3.7.1# download a model into the web cache,# assign its path to the variable "path"\download path https://modelhost.com/somemodel.safetensors# download to the models folder in the current directory# the models folder will be created if it does not exist# if somemodel.safetensors already exists it will be reused# instead of being downloaded again\download path https://modelhost.com/somemodel.safetensors -o models/somemodel.safetensors# download into the folder without specifying a name# the name will be derived from the URL or content disposition# header from the http request, if you are not careful you may# end up with a file named in a way you were not expecting.# only use this if you know how the service you are downloading# from behaves in this regard\download path https://modelhost.com/somemodel.safetensors -o models/# download a model into the web cache an overwrite any cached model using -x\download path https://modelhost.com/somemodel.safetensors -x# Download to an explicit path without any cached file reuse# using the -x/--overwrite argument. In effect, always freshly# download the file\download path https://modelhost.com/somemodel.safetensors -o models/somemodel.safetensors -x\download path https://modelhost.com/somemodel.safetensors -o models/ -x

#! /usr/bin/env dgenerate --file#! dgenerate 3.7.1\set my_variable {{ download('https://modelhost.com/model.safetensors') }}\set my_variable {{ download('https://modelhost.com/model.safetensors', output='model.safetensors') }}\set my_variable {{ download('https://modelhost.com/model.safetensors', output='directory/') }}# you can also use any template function with \setp (python expression evaluation)\setp my_variable download('https://modelhost.com/model.safetensors')

#! /usr/bin/env dgenerate --file#! dgenerate 3.7.1# exit the process with return code 0, which indicates success\print "success"\exit

#! /usr/bin/env dgenerate --file#! dgenerate 3.7.1# exit the process with return code 1, which indicates an error\print "some error occurred"\exit 1

dgenerate --file my-config.dgen

# Pipecat my-config.dgen | dgenerate# Redirectiondgenerate < my-config.dgen

dgenerate < my-config.dgen

Get-Content my-config.dgen | dgenerate

# Pipecat my-animations-config.dgen | dgenerate --frame-start 0 --frame-end 10# Redirectiondgenerate --frame-start 0 --frame-end 10 < my-animations-config.dgen

dgenerate --frame-start 0 --frame-end 10 < my-animations-config.dgen

Get-Content my-animations-config.dgen | dgenerate --frame-start 0 --frame-end 10

# all injected args\print {{ quote(injected_args) }}# just the injected device\print {{ '--device '+injected_device if injected_device else '' }}# was -v/--verbose injected?\print {{ '-v' if injected_verbose else '' }}# plugin module paths injected with --plugin-modules\print {{ quote(injected_plugin_modules) if injected_plugin_modules else '' }}