콘텐츠로 이동

Stable Diffusion XL ControlNet

ControlNet은 더 진보된 Stable Diffusion XL (SDXL) 모델에도 적용될 수 있어, 조건 이미지로부터 정밀한 구조적 가이드라인을 받아 고해상도 이미지 생성을 가능하게 합니다. Optimum RBLN은 RBLN NPU를 위한 가속화된 SDXL ControlNet 파이프라인을 제공합니다.

지원하는 파이프라인

  • SDXL ControlNet을 사용한 텍스트-이미지 변환: SDXL 기반 모델을 사용하여 제어 이미지로 가이드되는 텍스트 프롬프트에서 고해상도 이미지 생성.
  • SDXL ControlNet을 사용한 이미지-이미지 변환: SDXL 기반 모델을 사용하여 텍스트 프롬프트와 제어 이미지를 기반으로 기존 이미지 수정.

중요: Guidance Scale에 따른 배치 크기 설정

배치 크기와 Guidance Scale (SDXL)

다른 SDXL 파이프라인과 마찬가지로, guidance_scale > 1.0으로 ControlNet SDXL 파이프라인을 사용하면 UNet 및 ControlNet 모델의 실제 배치 크기가 2배가 됩니다.

RBLNStableDiffusionXLControlNetPipelineConfig의 unet 및 controlnet 섹션에 지정된 batch_size가 예상 실행 시간 배치 크기(guidance_scale > 1.0인 경우 일반적으로 추론 배치 크기의 2배)와 일치하는지 확인하십시오. 이를 생략하면 파이프라인의 batch_size를 기준으로 자동으로 2배가 됩니다.

API 참조

Classes

RBLNStableDiffusionXLControlNetPipeline

Bases: RBLNDiffusionMixin, StableDiffusionXLControlNetPipeline

RBLN-accelerated implementation of Stable Diffusion XL pipeline with ControlNet for high-resolution guided text-to-image generation.

This pipeline compiles Stable Diffusion XL and ControlNet models to run efficiently on RBLN NPUs, enabling high-performance inference for generating high-quality images with precise structural control and enhanced detail preservation.

Functions

__call__(prompt=None, prompt_2=None, image=None, height=None, width=None, num_inference_steps=50, denoising_end=None, guidance_scale=5.0, negative_prompt=None, negative_prompt_2=None, num_images_per_prompt=1, eta=0.0, generator=None, latents=None, prompt_embeds=None, negative_prompt_embeds=None, pooled_prompt_embeds=None, negative_pooled_prompt_embeds=None, ip_adapter_image=None, ip_adapter_image_embeds=None, output_type='pil', return_dict=True, cross_attention_kwargs=None, controlnet_conditioning_scale=1.0, guess_mode=False, control_guidance_start=0.0, control_guidance_end=1.0, original_size=None, crops_coords_top_left=(0, 0), target_size=None, negative_original_size=None, negative_crops_coords_top_left=(0, 0), negative_target_size=None, clip_skip=None, callback_on_step_end=None, callback_on_step_end_tensor_inputs=['latents'], **kwargs)

The call function to the pipeline for generation.

Parameters:

Name Type Description Default
prompt `str` or `List[str]`, *optional*

The prompt or prompts to guide image generation. If not defined, you need to pass prompt_embeds.

 None
prompt_2 `str` or `List[str]`, *optional*

The prompt or prompts to be sent to tokenizer_2 and text_encoder_2. If not defined, prompt is used in both text-encoders.

 None
image (`torch.FloatTensor`, `PIL.Image.Image`, `np.ndarray`, `List[torch.FloatTensor]`, `List[PIL.Image.Image]`, `List[np.ndarray]`, 
1
`List[List[torch.FloatTensor]]`, `List[List[np.ndarray]]` or `List[List[PIL.Image.Image]]`):

The ControlNet input condition to provide guidance to the unet for generation. If the type is specified as torch.FloatTensor, it is passed to ControlNet as is. PIL.Image.Image can also be accepted as an image. The dimensions of the output image defaults to image's dimensions. If height and/or width are passed, image is resized accordingly. If multiple ControlNets are specified in init, images must be passed as a list such that each element of the list can be correctly batched for input to a single ControlNet.

 required
height `int`, *optional*, defaults to `self.unet.config.sample_size * self.vae_scale_factor`

The height in pixels of the generated image. Anything below 512 pixels won't work well for stabilityai/stable-diffusion-xl-base-1.0 and checkpoints that are not specifically fine-tuned on low resolutions.

 None
width `int`, *optional*, defaults to `self.unet.config.sample_size * self.vae_scale_factor`

The width in pixels of the generated image. Anything below 512 pixels won't work well for stabilityai/stable-diffusion-xl-base-1.0 and checkpoints that are not specifically fine-tuned on low resolutions.

 None
num_inference_steps `int`, *optional*, defaults to 50

The number of denoising steps. More denoising steps usually lead to a higher quality image at the expense of slower inference.

 50
denoising_end `float`, *optional*

When specified, determines the fraction (between 0.0 and 1.0) of the total denoising process to be completed before it is intentionally prematurely terminated. As a result, the returned sample will still retain a substantial amount of noise as determined by the discrete timesteps selected by the scheduler. The denoising_end parameter should ideally be utilized when this pipeline forms a part of a "Mixture of Denoisers" multi-pipeline setup, as elaborated in Refining the Image Output

 None
guidance_scale `float`, *optional*, defaults to 5.0

A higher guidance scale value encourages the model to generate images closely linked to the text prompt at the expense of lower image quality. Guidance scale is enabled when guidance_scale > 1.

 5.0
negative_prompt `str` or `List[str]`, *optional*

The prompt or prompts to guide what to not include in image generation. If not defined, you need to pass negative_prompt_embeds instead. Ignored when not using guidance (guidance_scale < 1).

 None
negative_prompt_2 `str` or `List[str]`, *optional*

The prompt or prompts to guide what to not include in image generation. This is sent to tokenizer_2 and text_encoder_2. If not defined, negative_prompt is used in both text-encoders.

 None
num_images_per_prompt `int`, *optional*, defaults to 1

The number of images to generate per prompt.

 1
eta `float`, *optional*, defaults to 0.0

Corresponds to parameter eta (η) from the DDIM paper. Only applies to the [~schedulers.DDIMScheduler], and is ignored in other schedulers.

 0.0
generator `torch.Generator` or `List[torch.Generator]`, *optional*

A torch.Generator to make generation deterministic.

 None
latents `torch.FloatTensor`, *optional*

Pre-generated noisy latents sampled from a Gaussian distribution, to be used as inputs for image generation. Can be used to tweak the same generation with different prompts. If not provided, a latents tensor is generated by sampling using the supplied random generator.

 None
prompt_embeds `torch.FloatTensor`, *optional*

Pre-generated text embeddings. Can be used to easily tweak text inputs (prompt weighting). If not provided, text embeddings are generated from the prompt input argument.

 None
negative_prompt_embeds `torch.FloatTensor`, *optional*

Pre-generated negative text embeddings. Can be used to easily tweak text inputs (prompt weighting). If not provided, negative_prompt_embeds are generated from the negative_prompt input argument.

 None
pooled_prompt_embeds `torch.FloatTensor`, *optional*

Pre-generated pooled text embeddings. Can be used to easily tweak text inputs (prompt weighting). If not provided, pooled text embeddings are generated from prompt input argument.

 None
negative_pooled_prompt_embeds `torch.FloatTensor`, *optional*

Pre-generated negative pooled text embeddings. Can be used to easily tweak text inputs (prompt weighting). If not provided, pooled negative_prompt_embeds are generated from negative_prompt input argument.

 None
ip_adapter_image Optional[PipelineImageInput]

(PipelineImageInput, optional): Optional image input to work with IP Adapters.

 None
ip_adapter_image_embeds `List[torch.FloatTensor]`, *optional*

Pre-generated image embeddings for IP-Adapter. It should be a list of length same as number of IP-adapters. Each element should be a tensor of shape (batch_size, num_images, emb_dim). It should contain the negative image embedding if do_classifier_free_guidance is set to True. If not provided, embeddings are computed from the ip_adapter_image input argument.

 None
output_type `str`, *optional*, defaults to `"pil"`

The output format of the generated image. Choose between PIL.Image or np.array.

 'pil'
return_dict `bool`, *optional*, defaults to `True`

Whether or not to return a [~pipelines.stable_diffusion.StableDiffusionPipelineOutput] instead of a plain tuple.

 True
cross_attention_kwargs `dict`, *optional*

A kwargs dictionary that if specified is passed along to the [AttentionProcessor] as defined in self.processor.

 None
controlnet_conditioning_scale `float` or `List[float]`, *optional*, defaults to 1.0

The outputs of the ControlNet are multiplied by controlnet_conditioning_scale before they are added to the residual in the original unet. If multiple ControlNets are specified in init, you can set the corresponding scale as a list.

 1.0
guess_mode `bool`, *optional*, defaults to `False`

The ControlNet encoder tries to recognize the content of the input image even if you remove all prompts. A guidance_scale value between 3.0 and 5.0 is recommended.

 False
control_guidance_start `float` or `List[float]`, *optional*, defaults to 0.0

The percentage of total steps at which the ControlNet starts applying.

 0.0
control_guidance_end `float` or `List[float]`, *optional*, defaults to 1.0

The percentage of total steps at which the ControlNet stops applying.

 1.0
original_size `Tuple[int]`, *optional*, defaults to (1024, 1024)

If original_size is not the same as target_size the image will appear to be down- or upsampled. original_size defaults to (height, width) if not specified. Part of SDXL's micro-conditioning as explained in section 2.2 of https://huggingface.co/papers/2307.01952.

 None
crops_coords_top_left `Tuple[int]`, *optional*, defaults to (0, 0)

crops_coords_top_left can be used to generate an image that appears to be "cropped" from the position crops_coords_top_left downwards. Favorable, well-centered images are usually achieved by setting 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.

 (0, 0)
target_size `Tuple[int]`, *optional*, defaults to (1024, 1024)

For most cases, target_size should be set to the desired height and width of the generated image. If not specified it will default to (height, width). Part of SDXL's micro-conditioning as explained in section 2.2 of https://huggingface.co/papers/2307.01952.

 None
negative_original_size `Tuple[int]`, *optional*, defaults to (1024, 1024)

To 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.

 None
negative_crops_coords_top_left `Tuple[int]`, *optional*, defaults to (0, 0)

To 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.

 (0, 0)
negative_target_size `Tuple[int]`, *optional*, defaults to (1024, 1024)

To negatively condition the generation process based on a target image resolution. It should be as same as the 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.

 None
clip_skip `int`, *optional*

Number of layers to be skipped from CLIP while computing the prompt embeddings. A value of 1 means that the output of the pre-final layer will be used for computing the prompt embeddings.

 None
callback_on_step_end `Callable`, *optional*

A function that calls at the end of each denoising steps during the inference. The function is called with the following arguments: callback_on_step_end(self: DiffusionPipeline, step: int, timestep: int, callback_kwargs: Dict). callback_kwargs will include a list of all tensors as specified by callback_on_step_end_tensor_inputs.

 None
callback_on_step_end_tensor_inputs `List`, *optional*

The list of tensor inputs for the callback_on_step_end function. The tensors specified in the list will be passed as callback_kwargs argument. You will only be able to include variables listed in the ._callback_tensor_inputs attribute of your pipeine class.

 ['latents']

Examples:

Returns:

Type Description

[~pipelines.stable_diffusion.StableDiffusionPipelineOutput] or tuple: If return_dict is True, [~pipelines.stable_diffusion.StableDiffusionPipelineOutput] is returned, otherwise a tuple is returned containing the output images.
from_pretrained(model_id, *, export=None, model_save_dir=None, rbln_config={}, lora_ids=None, lora_weights_names=None, lora_scales=None, **kwargs) classmethod

Load a pretrained diffusion pipeline from a model checkpoint, with optional compilation for RBLN NPUs.

This method has two distinct operating modes
  • When export=True: Takes a PyTorch-based diffusion model, compiles it for RBLN NPUs, and loads the compiled model
  • When export=False: Loads an already compiled RBLN model from model_id without recompilation

It supports various diffusion pipelines including Stable Diffusion, Kandinsky, ControlNet, and other diffusers-based models.

Parameters:

Name Type Description Default
model_id `str`

The model ID or path to the pretrained model to load. Can be either:

  • A model ID from the HuggingFace Hub
  • A local path to a saved model directory
 required
export bool

If True, takes a PyTorch model from model_id and compiles it for RBLN NPU execution. If False, loads an already compiled RBLN model from model_id without recompilation.

 None
model_save_dir Optional[PathLike]

Directory to save the compiled model artifacts. Only used when export=True. If not provided and export=True, a temporary directory is used.

 None
rbln_config Dict[str, Any]

Configuration options for RBLN compilation. Can include settings for specific submodules such as text_encoder, unet, and vae. Configuration can be tailored to the specific pipeline being compiled.

 {}
lora_ids Optional[Union[str, List[str]]]

LoRA adapter ID(s) to load and apply before compilation. LoRA weights are fused into the model weights during compilation. Only used when export=True.

 None
lora_weights_names Optional[Union[str, List[str]]]

Names of specific LoRA weight files to load, corresponding to lora_ids. Only used when export=True.

 None
lora_scales Optional[Union[float, List[float]]]

Scaling factor(s) to apply to the LoRA adapter(s). Only used when export=True.

 None
kwargs Any

Additional arguments to pass to the underlying diffusion pipeline constructor or the RBLN compilation process. These may include parameters specific to individual submodules or the particular diffusion pipeline being used.

 {}

Returns:

Type Description
RBLNDiffusionMixin

A compiled or loaded diffusion pipeline that can be used for inference on RBLN NPU. The returned object is an instance of the class that called this method, inheriting from RBLNDiffusionMixin.

Functions

Classes

RBLNStableDiffusionXLControlNetImg2ImgPipeline

Bases: RBLNDiffusionMixin, StableDiffusionXLControlNetImg2ImgPipeline

RBLN-accelerated implementation of Stable Diffusion XL pipeline with ControlNet for high-resolution guided image-to-image generation.

This pipeline compiles Stable Diffusion XL and ControlNet models to run efficiently on RBLN NPUs, enabling high-performance inference for transforming input images with precise structural control and enhanced quality preservation.

Functions

__call__(prompt=None, prompt_2=None, image=None, control_image=None, height=None, width=None, strength=0.8, num_inference_steps=50, guidance_scale=5.0, negative_prompt=None, negative_prompt_2=None, num_images_per_prompt=1, eta=0.0, generator=None, latents=None, prompt_embeds=None, negative_prompt_embeds=None, pooled_prompt_embeds=None, negative_pooled_prompt_embeds=None, ip_adapter_image=None, ip_adapter_image_embeds=None, output_type='pil', return_dict=True, cross_attention_kwargs=None, controlnet_conditioning_scale=0.8, guess_mode=False, control_guidance_start=0.0, control_guidance_end=1.0, original_size=None, crops_coords_top_left=(0, 0), target_size=None, negative_original_size=None, negative_crops_coords_top_left=(0, 0), negative_target_size=None, aesthetic_score=6.0, negative_aesthetic_score=2.5, clip_skip=None, callback_on_step_end=None, callback_on_step_end_tensor_inputs=['latents'], **kwargs)

Function invoked when calling the pipeline for generation.

Parameters:

Name Type Description Default
prompt `str` or `List[str]`, *optional*

The prompt or prompts to guide the image generation. If not defined, one has to pass prompt_embeds. instead.

 None
prompt_2 `str` or `List[str]`, *optional*

The prompt or prompts to be sent to the tokenizer_2 and text_encoder_2. If not defined, prompt is used in both text-encoders

 None
image (`torch.FloatTensor`, `PIL.Image.Image`, `np.ndarray`, `List[torch.FloatTensor]`, `List[PIL.Image.Image]`, `List[np.ndarray]`, 
1
`List[List[torch.FloatTensor]]`, `List[List[np.ndarray]]` or `List[List[PIL.Image.Image]]`):

The initial image will be used as the starting point for the image generation process. Can also accept image latents as image, if passing latents directly, it will not be encoded again.

 required
control_image (`torch.FloatTensor`, `PIL.Image.Image`, `np.ndarray`, `List[torch.FloatTensor]`, `List[PIL.Image.Image]`, `List[np.ndarray]`, 
1
`List[List[torch.FloatTensor]]`, `List[List[np.ndarray]]` or `List[List[PIL.Image.Image]]`):

The ControlNet input condition. ControlNet uses this input condition to generate guidance to Unet. If the type is specified as Torch.FloatTensor, it is passed to ControlNet as is. PIL.Image.Image can also be accepted as an image. The dimensions of the output image defaults to image's dimensions. If height and/or width are passed, image is resized according to them. If multiple ControlNets are specified in init, images must be passed as a list such that each element of the list can be correctly batched for input to a single controlnet.

 required
height `int`, *optional*, defaults to the size of control_image

The height in pixels of the generated image. Anything below 512 pixels won't work well for stabilityai/stable-diffusion-xl-base-1.0 and checkpoints that are not specifically fine-tuned on low resolutions.

 None
width `int`, *optional*, defaults to the size of control_image

The width in pixels of the generated image. Anything below 512 pixels won't work well for stabilityai/stable-diffusion-xl-base-1.0 and checkpoints that are not specifically fine-tuned on low resolutions.

 None
strength `float`, *optional*, defaults to 0.8

Indicates extent to transform the reference image. Must be between 0 and 1. image is used as a starting point and more noise is added the higher the strength. The number of denoising steps depends on the amount of noise initially added. When strength is 1, added noise is maximum and the denoising process runs for the full number of iterations specified in num_inference_steps. A value of 1 essentially ignores image.

 0.8
num_inference_steps `int`, *optional*, defaults to 50

The number of denoising steps. More denoising steps usually lead to a higher quality image at the expense of slower inference.

 50
guidance_scale `float`, *optional*, defaults to 7.5

Guidance scale as defined in Classifier-Free Diffusion Guidance. guidance_scale is defined as w of equation 2. of Imagen Paper. Guidance scale is enabled by setting guidance_scale > 1. Higher guidance scale encourages to generate images that are closely linked to the text prompt, usually at the expense of lower image quality.

 5.0
negative_prompt `str` or `List[str]`, *optional*

The prompt or prompts not to guide the image generation. If not defined, one has to pass negative_prompt_embeds instead. Ignored when not using guidance (i.e., ignored if guidance_scale is less than 1).

 None
negative_prompt_2 `str` or `List[str]`, *optional*

The prompt or prompts not to guide the image generation to be sent to tokenizer_2 and text_encoder_2. If not defined, negative_prompt is used in both text-encoders

 None
num_images_per_prompt `int`, *optional*, defaults to 1

The number of images to generate per prompt.

 1
eta `float`, *optional*, defaults to 0.0

Corresponds to parameter eta (η) in the DDIM paper: https://arxiv.org/abs/2010.02502. Only applies to [schedulers.DDIMScheduler], will be ignored for others.

 0.0
generator `torch.Generator` or `List[torch.Generator]`, *optional*

One or a list of torch generator(s) to make generation deterministic.

 None
latents `torch.FloatTensor`, *optional*

Pre-generated noisy latents, sampled from a Gaussian distribution, to be used as inputs for image generation. Can be used to tweak the same generation with different prompts. If not provided, a latents tensor will ge generated by sampling using the supplied random generator.

 None
prompt_embeds `torch.FloatTensor`, *optional*

Pre-generated text embeddings. Can be used to easily tweak text inputs, e.g. prompt weighting. If not provided, text embeddings will be generated from prompt input argument.

 None
negative_prompt_embeds `torch.FloatTensor`, *optional*

Pre-generated negative text embeddings. Can be used to easily tweak text inputs, e.g. prompt weighting. If not provided, negative_prompt_embeds will be generated from negative_prompt input argument.

 None
pooled_prompt_embeds `torch.FloatTensor`, *optional*

Pre-generated pooled text embeddings. Can be used to easily tweak text inputs, e.g. prompt weighting. If not provided, pooled text embeddings will be generated from prompt input argument.

 None
negative_pooled_prompt_embeds `torch.FloatTensor`, *optional*

Pre-generated negative pooled text embeddings. Can be used to easily tweak text inputs, e.g. prompt weighting. If not provided, pooled negative_prompt_embeds will be generated from negative_prompt input argument.

 None
ip_adapter_image Optional[PipelineImageInput]

(PipelineImageInput, optional): Optional image input to work with IP Adapters.

 None
ip_adapter_image_embeds `List[torch.FloatTensor]`, *optional*

Pre-generated image embeddings for IP-Adapter. It should be a list of length same as number of IP-adapters. Each element should be a tensor of shape (batch_size, num_images, emb_dim). It should contain the negative image embedding if do_classifier_free_guidance is set to True. If not provided, embeddings are computed from the ip_adapter_image input argument.

 None
output_type `str`, *optional*, defaults to `"pil"`

The output format of the generate image. Choose between PIL: PIL.Image.Image or np.array.

 'pil'
return_dict `bool`, *optional*, defaults to `True`

Whether or not to return a [~pipelines.stable_diffusion.StableDiffusionPipelineOutput] instead of a plain tuple.

 True
cross_attention_kwargs `dict`, *optional*

A kwargs dictionary that if specified is passed along to the AttentionProcessor as defined under self.processor in diffusers.models.attention_processor.

 None
controlnet_conditioning_scale `float` or `List[float]`, *optional*, defaults to 1.0

The outputs of the controlnet are multiplied by controlnet_conditioning_scale before they are added to the residual in the original unet. If multiple ControlNets are specified in init, you can set the corresponding scale as a list.

 0.8
guess_mode `bool`, *optional*, defaults to `False`

In this mode, the ControlNet encoder will try best to recognize the content of the input image even if you remove all prompts. The guidance_scale between 3.0 and 5.0 is recommended.

 False
control_guidance_start `float` or `List[float]`, *optional*, defaults to 0.0

The percentage of total steps at which the controlnet starts applying.

 0.0
control_guidance_end `float` or `List[float]`, *optional*, defaults to 1.0

The percentage of total steps at which the controlnet stops applying.

 1.0
original_size `Tuple[int]`, *optional*, defaults to (1024, 1024)

If original_size is not the same as target_size the image will appear to be down- or upsampled. original_size defaults to (height, width) if not specified. Part of SDXL's micro-conditioning as explained in section 2.2 of https://huggingface.co/papers/2307.01952.

 None
crops_coords_top_left `Tuple[int]`, *optional*, defaults to (0, 0)

crops_coords_top_left can be used to generate an image that appears to be "cropped" from the position crops_coords_top_left downwards. Favorable, well-centered images are usually achieved by setting 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.

 (0, 0)
target_size `Tuple[int]`, *optional*, defaults to (1024, 1024)

For most cases, target_size should be set to the desired height and width of the generated image. If not specified it will default to (height, width). Part of SDXL's micro-conditioning as explained in section 2.2 of https://huggingface.co/papers/2307.01952.

 None
negative_original_size `Tuple[int]`, *optional*, defaults to (1024, 1024)

To 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.

 None
negative_crops_coords_top_left `Tuple[int]`, *optional*, defaults to (0, 0)

To 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.

 (0, 0)
negative_target_size `Tuple[int]`, *optional*, defaults to (1024, 1024)

To negatively condition the generation process based on a target image resolution. It should be as same as the 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.

 None
aesthetic_score `float`, *optional*, defaults to 6.0

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.

 6.0
negative_aesthetic_score `float`, *optional*, defaults to 2.5

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.

 2.5
clip_skip `int`, *optional*

Number of layers to be skipped from CLIP while computing the prompt embeddings. A value of 1 means that the output of the pre-final layer will be used for computing the prompt embeddings.

 None
callback_on_step_end `Callable`, *optional*

A function that calls at the end of each denoising steps during the inference. The function is called with the following arguments: callback_on_step_end(self: DiffusionPipeline, step: int, timestep: int, callback_kwargs: Dict). callback_kwargs will include a list of all tensors as specified by callback_on_step_end_tensor_inputs.

 None
callback_on_step_end_tensor_inputs `List`, *optional*

The list of tensor inputs for the callback_on_step_end function. The tensors specified in the list will be passed as callback_kwargs argument. You will only be able to include variables listed in the ._callback_tensor_inputs attribute of your pipeine class.

 ['latents']

Examples:

Returns:

Type Description

[~pipelines.stable_diffusion.StableDiffusionPipelineOutput] or tuple:

[~pipelines.stable_diffusion.StableDiffusionPipelineOutput] if return_dict is True, otherwise a tuple

containing the output images.
from_pretrained(model_id, *, export=None, model_save_dir=None, rbln_config={}, lora_ids=None, lora_weights_names=None, lora_scales=None, **kwargs) classmethod

Load a pretrained diffusion pipeline from a model checkpoint, with optional compilation for RBLN NPUs.

This method has two distinct operating modes
  • When export=True: Takes a PyTorch-based diffusion model, compiles it for RBLN NPUs, and loads the compiled model
  • When export=False: Loads an already compiled RBLN model from model_id without recompilation

It supports various diffusion pipelines including Stable Diffusion, Kandinsky, ControlNet, and other diffusers-based models.

Parameters:

Name Type Description Default
model_id `str`

The model ID or path to the pretrained model to load. Can be either:

  • A model ID from the HuggingFace Hub
  • A local path to a saved model directory
 required
export bool

If True, takes a PyTorch model from model_id and compiles it for RBLN NPU execution. If False, loads an already compiled RBLN model from model_id without recompilation.

 None
model_save_dir Optional[PathLike]

Directory to save the compiled model artifacts. Only used when export=True. If not provided and export=True, a temporary directory is used.

 None
rbln_config Dict[str, Any]

Configuration options for RBLN compilation. Can include settings for specific submodules such as text_encoder, unet, and vae. Configuration can be tailored to the specific pipeline being compiled.

 {}
lora_ids Optional[Union[str, List[str]]]

LoRA adapter ID(s) to load and apply before compilation. LoRA weights are fused into the model weights during compilation. Only used when export=True.

 None
lora_weights_names Optional[Union[str, List[str]]]

Names of specific LoRA weight files to load, corresponding to lora_ids. Only used when export=True.

 None
lora_scales Optional[Union[float, List[float]]]

Scaling factor(s) to apply to the LoRA adapter(s). Only used when export=True.

 None
kwargs Any

Additional arguments to pass to the underlying diffusion pipeline constructor or the RBLN compilation process. These may include parameters specific to individual submodules or the particular diffusion pipeline being used.

 {}

Returns:

Type Description
RBLNDiffusionMixin

A compiled or loaded diffusion pipeline that can be used for inference on RBLN NPU. The returned object is an instance of the class that called this method, inheriting from RBLNDiffusionMixin.

Functions

Classes

RBLNStableDiffusionXLControlNetPipelineBaseConfig

Bases: RBLNModelConfig

Base configuration for Stable Diffusion XL ControlNet pipelines.

Functions

__init__(text_encoder=None, text_encoder_2=None, unet=None, vae=None, controlnet=None, *, batch_size=None, img_height=None, img_width=None, height=None, width=None, sample_size=None, image_size=None, guidance_scale=None, **kwargs)

Parameters:

Name Type Description Default
text_encoder Optional[RBLNCLIPTextModelConfig]

Configuration for the primary text encoder. Initialized as RBLNCLIPTextModelConfig if not provided.

 None
text_encoder_2 Optional[RBLNCLIPTextModelWithProjectionConfig]

Configuration for the secondary text encoder. Initialized as RBLNCLIPTextModelWithProjectionConfig if not provided.

 None
unet Optional[RBLNUNet2DConditionModelConfig]

Configuration for the UNet model component. Initialized as RBLNUNet2DConditionModelConfig if not provided.

 None
vae Optional[RBLNAutoencoderKLConfig]

Configuration for the VAE model component. Initialized as RBLNAutoencoderKLConfig if not provided.

 None
controlnet Optional[RBLNControlNetModelConfig]

Configuration for the ControlNet model component. Initialized as RBLNControlNetModelConfig if not provided.

 None
batch_size Optional[int]

Batch size for inference, applied to all submodules.

 None
img_height Optional[int]

Height of the generated images.

 None
img_width Optional[int]

Width of the generated images.

 None
height Optional[int]

Height of the generated images.

 None
width Optional[int]

Width of the generated images.

 None
sample_size Optional[Tuple[int, int]]

Spatial dimensions for the UNet model.

 None
image_size Optional[Tuple[int, int]]

Alternative way to specify image dimensions. Cannot be used together with img_height/img_width.

 None
guidance_scale Optional[float]

Scale for classifier-free guidance.

 None
kwargs Any

Additional arguments passed to the parent RBLNModelConfig.

 {}

Raises:

Type Description
ValueError

If both image_size and img_height/img_width are provided.
Note

When guidance_scale > 1.0, the UNet batch size is automatically doubled to accommodate classifier-free guidance.

load(path, **kwargs) classmethod

Load a RBLNModelConfig from a path.

Parameters:

Name Type Description Default
path str

Path to the RBLNModelConfig file or directory containing the config file.

 required
kwargs Any

Additional keyword arguments to override configuration values. Keys starting with 'rbln_' will have the prefix removed and be used to update the configuration.

 {}

Returns:

Name Type Description
RBLNModelConfig RBLNModelConfig

The loaded configuration instance.
Note

This method loads the configuration from the specified path and applies any provided overrides. If the loaded configuration class doesn't match the expected class, a warning will be logged.

RBLNStableDiffusionXLControlNetPipelineConfig

Bases: RBLNStableDiffusionXLControlNetPipelineBaseConfig

Configuration for Stable Diffusion XL ControlNet pipeline.

Functions

__init__(text_encoder=None, text_encoder_2=None, unet=None, vae=None, controlnet=None, *, batch_size=None, img_height=None, img_width=None, height=None, width=None, sample_size=None, image_size=None, guidance_scale=None, **kwargs)

Parameters:

Name Type Description Default
text_encoder Optional[RBLNCLIPTextModelConfig]

Configuration for the primary text encoder. Initialized as RBLNCLIPTextModelConfig if not provided.

 None
text_encoder_2 Optional[RBLNCLIPTextModelWithProjectionConfig]

Configuration for the secondary text encoder. Initialized as RBLNCLIPTextModelWithProjectionConfig if not provided.

 None
unet Optional[RBLNUNet2DConditionModelConfig]

Configuration for the UNet model component. Initialized as RBLNUNet2DConditionModelConfig if not provided.

 None
vae Optional[RBLNAutoencoderKLConfig]

Configuration for the VAE model component. Initialized as RBLNAutoencoderKLConfig if not provided.

 None
controlnet Optional[RBLNControlNetModelConfig]

Configuration for the ControlNet model component. Initialized as RBLNControlNetModelConfig if not provided.

 None
batch_size Optional[int]

Batch size for inference, applied to all submodules.

 None
img_height Optional[int]

Height of the generated images.

 None
img_width Optional[int]

Width of the generated images.

 None
height Optional[int]

Height of the generated images.

 None
width Optional[int]

Width of the generated images.

 None
sample_size Optional[Tuple[int, int]]

Spatial dimensions for the UNet model.

 None
image_size Optional[Tuple[int, int]]

Alternative way to specify image dimensions. Cannot be used together with img_height/img_width.

 None
guidance_scale Optional[float]

Scale for classifier-free guidance.

 None
kwargs Any

Additional arguments passed to the parent RBLNModelConfig.

 {}

Raises:

Type Description
ValueError

If both image_size and img_height/img_width are provided.
Note

When guidance_scale > 1.0, the UNet batch size is automatically doubled to accommodate classifier-free guidance.

load(path, **kwargs) classmethod

Load a RBLNModelConfig from a path.

Parameters:

Name Type Description Default
path str

Path to the RBLNModelConfig file or directory containing the config file.

 required
kwargs Any

Additional keyword arguments to override configuration values. Keys starting with 'rbln_' will have the prefix removed and be used to update the configuration.

 {}

Returns:

Name Type Description
RBLNModelConfig RBLNModelConfig

The loaded configuration instance.
Note

This method loads the configuration from the specified path and applies any provided overrides. If the loaded configuration class doesn't match the expected class, a warning will be logged.

RBLNStableDiffusionXLControlNetImg2ImgPipelineConfig

Bases: RBLNStableDiffusionXLControlNetPipelineBaseConfig

Configuration for Stable Diffusion XL ControlNet image-to-image pipeline.

Functions

__init__(text_encoder=None, text_encoder_2=None, unet=None, vae=None, controlnet=None, *, batch_size=None, img_height=None, img_width=None, height=None, width=None, sample_size=None, image_size=None, guidance_scale=None, **kwargs)

Parameters:

Name Type Description Default
text_encoder Optional[RBLNCLIPTextModelConfig]

Configuration for the primary text encoder. Initialized as RBLNCLIPTextModelConfig if not provided.

 None
text_encoder_2 Optional[RBLNCLIPTextModelWithProjectionConfig]

Configuration for the secondary text encoder. Initialized as RBLNCLIPTextModelWithProjectionConfig if not provided.

 None
unet Optional[RBLNUNet2DConditionModelConfig]

Configuration for the UNet model component. Initialized as RBLNUNet2DConditionModelConfig if not provided.

 None
vae Optional[RBLNAutoencoderKLConfig]

Configuration for the VAE model component. Initialized as RBLNAutoencoderKLConfig if not provided.

 None
controlnet Optional[RBLNControlNetModelConfig]

Configuration for the ControlNet model component. Initialized as RBLNControlNetModelConfig if not provided.

 None
batch_size Optional[int]

Batch size for inference, applied to all submodules.

 None
img_height Optional[int]

Height of the generated images.

 None
img_width Optional[int]

Width of the generated images.

 None
height Optional[int]

Height of the generated images.

 None
width Optional[int]

Width of the generated images.

 None
sample_size Optional[Tuple[int, int]]

Spatial dimensions for the UNet model.

 None
image_size Optional[Tuple[int, int]]

Alternative way to specify image dimensions. Cannot be used together with img_height/img_width.

 None
guidance_scale Optional[float]

Scale for classifier-free guidance.

 None
kwargs Any

Additional arguments passed to the parent RBLNModelConfig.

 {}

Raises:

Type Description
ValueError

If both image_size and img_height/img_width are provided.
Note

When guidance_scale > 1.0, the UNet batch size is automatically doubled to accommodate classifier-free guidance.

load(path, **kwargs) classmethod

Load a RBLNModelConfig from a path.

Parameters:

Name Type Description Default
path str

Path to the RBLNModelConfig file or directory containing the config file.

 required
kwargs Any

Additional keyword arguments to override configuration values. Keys starting with 'rbln_' will have the prefix removed and be used to update the configuration.

 {}

Returns:

Name Type Description
RBLNModelConfig RBLNModelConfig

The loaded configuration instance.
Note

This method loads the configuration from the specified path and applies any provided overrides. If the loaded configuration class doesn't match the expected class, a warning will be logged.