Kandinsky V2.2¶

Kandinsky V2.2는 텍스트-이미지 잠재 확산 모델입니다. RBLN Optimum을 사용하면 RBLN NPU에서 Kandinsky V2.2 파이프라인을 가속할 수 있습니다.

지원 NPU

Diffusers 파이프라인은 ATOM™에서만 실행됩니다. Rebel100™은 아직 지원하지 않습니다.

지원하는 파이프라인¶

RBLN Optimum은 여러 Kandinsky V2.2 파이프라인을 지원합니다:

파이프라인	클래스	설명
텍스트-이미지 변환(Text-to-Image)	`RBLNKandinskyV22CombinedPipeline`	텍스트 프롬프트에서 이미지 생성(Prior + Decoder 사용).
이미지-이미지 변환(Image-to-Image)	`RBLNKandinskyV22Img2ImgCombinedPipeline`	텍스트 프롬프트를 기반으로 기존 이미지 수정(Prior + Img2Img Decoder 사용).
인페인팅(Inpainting)	`RBLNKandinskyV22InpaintCombinedPipeline`	텍스트 프롬프트에 따라 이미지의 마스킹된 영역 채우기(Prior + Inpaint Decoder 사용).
Prior	`RBLNKandinskyV22PriorPipeline`	Decoder 파이프라인이 사용할 이미지 임베딩 생성.
Decoder	`RBLNKandinskyV22Pipeline`, `RBLNKandinskyV22Img2ImgPipeline`, `RBLNKandinskyV22InpaintPipeline`	Prior가 만든 임베딩에서 이미지 생성. 작업별로 클래스가 하나씩 있습니다.

중요: guidance scale을 위한 배치 크기 설정¶

배치 크기와 guidance scale

Kandinsky V2.2를 guidance scale > 1.0(기본값)으로 사용할 때, UNet과 Prior 두 구성 요소의 실제 배치 크기는 classifier-free guidance가 적용되어 런타임 중에 두 배가 됩니다.

RBLN NPU는 정적 그래프 컴파일을 사용하기 때문에, 이러한 구성 요소의 컴파일 시 배치 크기가 런타임 배치 크기와 일치해야 합니다. 그렇지 않으면 컴파일된 그래프를 실행할 수 없어 추론이 실패합니다.

기본 동작¶

UNet이나 Prior의 배치 크기를 명시적으로 지정하지 않으면 RBLN Optimum은 다음과 같이 동작합니다:

기본 guidance scale(1.0보다 큼)을 사용할 것으로 가정
UNet과 Prior의 배치 크기를 파이프라인 배치 크기의 2배로 자동 설정

기본 guidance scale을 사용할 계획이라면 이 자동 설정이 정상적으로 작동합니다. 그러나 다른 guidance scale을 사용하거나 더 많은 제어가 필요한 경우, 배치 크기를 명시적으로 설정해야 합니다.

예제: 명시적으로 배치 크기 설정하기(guidance scale = 1.0)¶

정확히 guidance scale 1.0을 사용할 계획이라면(classifier-free guidance를 사용하지 않음), 배치 크기를 추론 배치 크기와 일치하도록 명시적으로 설정합니다:

사용 예제¶

방법 1: 개별 Prior 및 Decoder 파이프라인 사용¶

이 방식에서는 중간 이미지 임베딩을 직접 다룰 수 있습니다:

import torch
from optimum.rbln import RBLNKandinskyV22PriorPipeline, RBLNKandinskyV22Pipeline

# Prior 파이프라인 로드 및 컴파일
prior_pipe = RBLNKandinskyV22PriorPipeline.from_pretrained(
    "kandinsky-community/kandinsky-2-2-prior",
    export=True,  # 첫 번째 컴파일 시 True로 설정
)

# 특정 이미지 크기로 Decoder 파이프라인 로드 및 컴파일
pipe = RBLNKandinskyV22Pipeline.from_pretrained(
    "kandinsky-community/kandinsky-2-2-decoder",
    export=True,
    rbln_img_height=768,
    rbln_img_width=768,
)

# Prior를 사용하여 텍스트 프롬프트에서 이미지 임베딩 생성
prompt = "a cute cat sitting on a beach, 4k photo"
generator = torch.manual_seed(42)  # 재현 가능한 결과를 위해
prior_output = prior_pipe(prompt, generator=generator)
image_emb = prior_output.image_embeds
negative_image_emb = prior_output.negative_image_embeds

# Decoder와 이미지 임베딩을 사용하여 최종 이미지 생성
output = pipe(
    image_embeds=image_emb,
    negative_image_embeds=negative_image_emb,
    height=768,
    width=768,
    num_inference_steps=50,
    generator=generator,
)

# 생성된 이미지 저장
image = output.images[0]
image.save("generated_cat.png")

방법 2: 결합 파이프라인 사용¶

결합 파이프라인은 Prior와 Decoder를 한 번의 호출로 실행합니다:

import torch
from optimum.rbln import RBLNKandinskyV22CombinedPipeline

# 결합 파이프라인 로드 및 컴파일
pipe = RBLNKandinskyV22CombinedPipeline.from_pretrained(
    "kandinsky-community/kandinsky-2-2-decoder",  # 주 디코더 모델
    export=True,  # 첫 번째 컴파일 시 True로 설정
    rbln_img_height=768,
    rbln_img_width=768,
)

# 이미지 생성 및 저장
prompt = "a cute cat sitting on a beach, 4k photo"
generator = torch.manual_seed(42)  # 재현 가능한 결과를 위해
image = pipe(
    prompt, height=768, width=768, num_inference_steps=50, generator=generator
).images[0]

image.save("generated_cat.png")

API 참조¶

`RBLNKandinskyV22Pipeline` ¶

Bases: RBLNDiffusionMixin, KandinskyV22Pipeline

RBLN-accelerated implementation of Kandinsky 2.2 pipeline for text-to-image generation.

This pipeline compiles Kandinsky 2.2 models to run efficiently on RBLN NPUs, enabling high-performance inference for generating images with distinctive artistic style and enhanced visual quality.

`from_pretrained(model_id, *, export=None, model_save_dir=None, rbln_config=None, 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 \| None`	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`	`str \| PathLike[str] \| None`	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] \| RBLNModelConfig \| None`	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.	`None`
`lora_ids`	`str \| list[str] \| None`	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`	`str \| list[str] \| None`	Names of specific LoRA weight files to load, corresponding to lora_ids. Only used when `export=True`.	`None`
`lora_scales`	`float \| list[float] \| None`	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
`Self`	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.

`RBLNKandinskyV22PriorPipeline` ¶

Bases: RBLNDiffusionMixin, KandinskyV22PriorPipeline

RBLN-accelerated implementation of Kandinsky 2.2 prior pipeline for text and image embedding generation.

This pipeline compiles Kandinsky 2.2 prior models to run efficiently on RBLN NPUs, enabling high-performance inference for generating image embeddings from text prompts and image inputs for downstream generation tasks.

`from_pretrained(model_id, *, export=None, model_save_dir=None, rbln_config=None, 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 \| None`	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`	`str \| PathLike[str] \| None`	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] \| RBLNModelConfig \| None`	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.	`None`
`lora_ids`	`str \| list[str] \| None`	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`	`str \| list[str] \| None`	Names of specific LoRA weight files to load, corresponding to lora_ids. Only used when `export=True`.	`None`
`lora_scales`	`float \| list[float] \| None`	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
`Self`	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.

`RBLNKandinskyV22Img2ImgPipeline` ¶

Bases: RBLNDiffusionMixin, KandinskyV22Img2ImgPipeline

RBLN-accelerated implementation of Kandinsky 2.2 pipeline for image-to-image generation.

This pipeline compiles Kandinsky 2.2 models to run efficiently on RBLN NPUs, enabling high-performance inference for transforming input images with distinctive artistic style and enhanced visual fidelity.

`from_pretrained(model_id, *, export=None, model_save_dir=None, rbln_config=None, 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 \| None`	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`	`str \| PathLike[str] \| None`	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] \| RBLNModelConfig \| None`	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.	`None`
`lora_ids`	`str \| list[str] \| None`	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`	`str \| list[str] \| None`	Names of specific LoRA weight files to load, corresponding to lora_ids. Only used when `export=True`.	`None`
`lora_scales`	`float \| list[float] \| None`	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
`Self`	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.

`RBLNKandinskyV22InpaintPipeline` ¶

Bases: RBLNDiffusionMixin, KandinskyV22InpaintPipeline

RBLN-accelerated implementation of Kandinsky 2.2 pipeline for image inpainting.

This pipeline compiles Kandinsky 2.2 models to run efficiently on RBLN NPUs, enabling high-performance inference for filling masked regions with distinctive artistic style and seamless content integration.

`from_pretrained(model_id, *, export=None, model_save_dir=None, rbln_config=None, 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 \| None`	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`	`str \| PathLike[str] \| None`	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] \| RBLNModelConfig \| None`	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.	`None`
`lora_ids`	`str \| list[str] \| None`	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`	`str \| list[str] \| None`	Names of specific LoRA weight files to load, corresponding to lora_ids. Only used when `export=True`.	`None`
`lora_scales`	`float \| list[float] \| None`	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
`Self`	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.

`RBLNKandinskyV22CombinedPipeline` ¶

Bases: RBLNDiffusionMixin, KandinskyV22CombinedPipeline

RBLN-accelerated implementation of Kandinsky 2.2 combined pipeline for end-to-end text-to-image generation.

This pipeline compiles both prior and decoder Kandinsky 2.2 models to run efficiently on RBLN NPUs, enabling high-performance inference for complete text-to-image generation with distinctive artistic style.

`from_pretrained(model_id, *, export=None, model_save_dir=None, rbln_config=None, 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 \| None`	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`	`str \| PathLike[str] \| None`	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] \| RBLNModelConfig \| None`	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.	`None`
`lora_ids`	`str \| list[str] \| None`	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`	`str \| list[str] \| None`	Names of specific LoRA weight files to load, corresponding to lora_ids. Only used when `export=True`.	`None`
`lora_scales`	`float \| list[float] \| None`	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
`Self`	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.

`RBLNKandinskyV22Img2ImgCombinedPipeline` ¶

Bases: RBLNDiffusionMixin, KandinskyV22Img2ImgCombinedPipeline

RBLN-accelerated implementation of Kandinsky 2.2 combined pipeline for end-to-end image-to-image generation.

This pipeline compiles both prior and decoder Kandinsky 2.2 models to run efficiently on RBLN NPUs, enabling high-performance inference for complete image-to-image transformation with distinctive artistic style.

`from_pretrained(model_id, *, export=None, model_save_dir=None, rbln_config=None, 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 \| None`	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`	`str \| PathLike[str] \| None`	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] \| RBLNModelConfig \| None`	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.	`None`
`lora_ids`	`str \| list[str] \| None`	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`	`str \| list[str] \| None`	Names of specific LoRA weight files to load, corresponding to lora_ids. Only used when `export=True`.	`None`
`lora_scales`	`float \| list[float] \| None`	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
`Self`	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.

`RBLNKandinskyV22InpaintCombinedPipeline` ¶

Bases: RBLNDiffusionMixin, KandinskyV22InpaintCombinedPipeline

RBLN-accelerated implementation of Kandinsky 2.2 combined pipeline for end-to-end image inpainting.

This pipeline compiles both prior and decoder Kandinsky 2.2 models to run efficiently on RBLN NPUs, enabling high-performance inference for complete image inpainting with distinctive artistic style and seamless integration.

`from_pretrained(model_id, *, export=None, model_save_dir=None, rbln_config=None, 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 \| None`	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`	`str \| PathLike[str] \| None`	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] \| RBLNModelConfig \| None`	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.	`None`
`lora_ids`	`str \| list[str] \| None`	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`	`str \| list[str] \| None`	Names of specific LoRA weight files to load, corresponding to lora_ids. Only used when `export=True`.	`None`
`lora_scales`	`float \| list[float] \| None`	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
`Self`	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.

`RBLNKandinskyV22PipelineBaseConfig` ¶

Bases: RBLNModelConfig

`init(unet=None, movq=None, *, sample_size=None, batch_size=None, guidance_scale=None, image_size=None, img_height=None, img_width=None, height=None, width=None, **kwargs)` ¶

Parameters:

Name	Type	Description	Default
`unet`	`RBLNUNet2DConditionModelConfig \| None`	Configuration for the UNet model component. Initialized as RBLNUNet2DConditionModelConfig if not provided.	`None`
`movq`	`RBLNVQModelConfig \| None`	Configuration for the MoVQ (VQ-GAN) model component. Initialized as RBLNVQModelConfig if not provided.	`None`
`sample_size`	`tuple[int, int] \| None`	Spatial dimensions for the UNet model.	`None`
`batch_size`	`int \| None`	Batch size for inference, applied to all submodules.	`None`
`guidance_scale`	`float \| None`	Scale for classifier-free guidance.	`None`
`image_size`	`tuple[int, int] \| None`	Dimensions for the generated images. Cannot be used together with img_height/img_width.	`None`
`img_height`	`int \| None`	Height of the generated images.	`None`
`img_width`	`int \| None`	Width of the generated images.	`None`
`height`	`int \| None`	Height of the generated images.	`None`
`width`	`int \| None`	Width of the generated images.	`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.

`from_pretrained(path, rbln_config=None, return_unused_kwargs=False, **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
`rbln_config`	`dict[str, Any] \| None`	Additional configuration to override.	`None`
`return_unused_kwargs`	`bool`	Whether to return unused kwargs.	`False`
`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`	`Self \| tuple[Self, dict[str, Any]]`	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.

Examples:

config = RBLNResNetForImageClassificationConfig.from_pretrained("/path/to/model")

`initialize_submodule_config(submodule_config=None, force_kwargs=False, **kwargs)` ¶

Build the config of a submodule from the parent's runtime options and kwargs.

Returns the submodule config instance when its class is known (an instance was passed, or cls_name was given), otherwise the merged kwargs dict for the model class to resolve later.

`get_load_overrides()` ¶

Extract the options that must cross the load boundary, for passing to from_pretrained.

Returns the explicitly-set runtime options (device, device_map, create_runtimes, activate_profiler, timeout) and the subclass-specific non-save attributes (load-behavior flags such as _load_visual_runtime, which are never serialized so they can only travel with the caller), recursively including submodules. Compile-time attributes are not included: when loading a compiled model they come from the artifact's rbln_config.json, and this dict is merged on top of it.

Returns:

Type	Description
`dict[str, Any]`	Dictionary of load overrides, keyed like the `rbln_config` dict accepted by `from_pretrained`.

`RBLNKandinskyV22PipelineConfig` ¶

Bases: RBLNKandinskyV22PipelineBaseConfig

Configuration class for the Kandinsky V2.2 text-to-image decoder pipeline.

`initialize_submodule_config(submodule_config=None, force_kwargs=False, **kwargs)` ¶

Build the config of a submodule from the parent's runtime options and kwargs.

Returns the submodule config instance when its class is known (an instance was passed, or cls_name was given), otherwise the merged kwargs dict for the model class to resolve later.

`init(unet=None, movq=None, *, sample_size=None, batch_size=None, guidance_scale=None, image_size=None, img_height=None, img_width=None, height=None, width=None, **kwargs)` ¶

Parameters:

Name	Type	Description	Default
`unet`	`RBLNUNet2DConditionModelConfig \| None`	Configuration for the UNet model component. Initialized as RBLNUNet2DConditionModelConfig if not provided.	`None`
`movq`	`RBLNVQModelConfig \| None`	Configuration for the MoVQ (VQ-GAN) model component. Initialized as RBLNVQModelConfig if not provided.	`None`
`sample_size`	`tuple[int, int] \| None`	Spatial dimensions for the UNet model.	`None`
`batch_size`	`int \| None`	Batch size for inference, applied to all submodules.	`None`
`guidance_scale`	`float \| None`	Scale for classifier-free guidance.	`None`
`image_size`	`tuple[int, int] \| None`	Dimensions for the generated images. Cannot be used together with img_height/img_width.	`None`
`img_height`	`int \| None`	Height of the generated images.	`None`
`img_width`	`int \| None`	Width of the generated images.	`None`
`height`	`int \| None`	Height of the generated images.	`None`
`width`	`int \| None`	Width of the generated images.	`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.

`get_load_overrides()` ¶

Extract the options that must cross the load boundary, for passing to from_pretrained.

Returns the explicitly-set runtime options (device, device_map, create_runtimes, activate_profiler, timeout) and the subclass-specific non-save attributes (load-behavior flags such as _load_visual_runtime, which are never serialized so they can only travel with the caller), recursively including submodules. Compile-time attributes are not included: when loading a compiled model they come from the artifact's rbln_config.json, and this dict is merged on top of it.

Returns:

Type	Description
`dict[str, Any]`	Dictionary of load overrides, keyed like the `rbln_config` dict accepted by `from_pretrained`.

`from_pretrained(path, rbln_config=None, return_unused_kwargs=False, **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
`rbln_config`	`dict[str, Any] \| None`	Additional configuration to override.	`None`
`return_unused_kwargs`	`bool`	Whether to return unused kwargs.	`False`
`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`	`Self \| tuple[Self, dict[str, Any]]`	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.

Examples:

config = RBLNResNetForImageClassificationConfig.from_pretrained("/path/to/model")

`RBLNKandinskyV22Img2ImgPipelineConfig` ¶

Bases: RBLNKandinskyV22PipelineBaseConfig

Configuration class for the Kandinsky V2.2 image-to-image decoder pipeline.

`initialize_submodule_config(submodule_config=None, force_kwargs=False, **kwargs)` ¶

Build the config of a submodule from the parent's runtime options and kwargs.

Returns the submodule config instance when its class is known (an instance was passed, or cls_name was given), otherwise the merged kwargs dict for the model class to resolve later.

`init(unet=None, movq=None, *, sample_size=None, batch_size=None, guidance_scale=None, image_size=None, img_height=None, img_width=None, height=None, width=None, **kwargs)` ¶

Parameters:

Name	Type	Description	Default
`unet`	`RBLNUNet2DConditionModelConfig \| None`	Configuration for the UNet model component. Initialized as RBLNUNet2DConditionModelConfig if not provided.	`None`
`movq`	`RBLNVQModelConfig \| None`	Configuration for the MoVQ (VQ-GAN) model component. Initialized as RBLNVQModelConfig if not provided.	`None`
`sample_size`	`tuple[int, int] \| None`	Spatial dimensions for the UNet model.	`None`
`batch_size`	`int \| None`	Batch size for inference, applied to all submodules.	`None`
`guidance_scale`	`float \| None`	Scale for classifier-free guidance.	`None`
`image_size`	`tuple[int, int] \| None`	Dimensions for the generated images. Cannot be used together with img_height/img_width.	`None`
`img_height`	`int \| None`	Height of the generated images.	`None`
`img_width`	`int \| None`	Width of the generated images.	`None`
`height`	`int \| None`	Height of the generated images.	`None`
`width`	`int \| None`	Width of the generated images.	`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.

`get_load_overrides()` ¶

Extract the options that must cross the load boundary, for passing to from_pretrained.

Returns the explicitly-set runtime options (device, device_map, create_runtimes, activate_profiler, timeout) and the subclass-specific non-save attributes (load-behavior flags such as _load_visual_runtime, which are never serialized so they can only travel with the caller), recursively including submodules. Compile-time attributes are not included: when loading a compiled model they come from the artifact's rbln_config.json, and this dict is merged on top of it.

Returns:

Type	Description
`dict[str, Any]`	Dictionary of load overrides, keyed like the `rbln_config` dict accepted by `from_pretrained`.

`from_pretrained(path, rbln_config=None, return_unused_kwargs=False, **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
`rbln_config`	`dict[str, Any] \| None`	Additional configuration to override.	`None`
`return_unused_kwargs`	`bool`	Whether to return unused kwargs.	`False`
`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`	`Self \| tuple[Self, dict[str, Any]]`	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.

Examples:

config = RBLNResNetForImageClassificationConfig.from_pretrained("/path/to/model")

`RBLNKandinskyV22InpaintPipelineConfig` ¶

Bases: RBLNKandinskyV22PipelineBaseConfig

Configuration class for the Kandinsky V2.2 inpainting decoder pipeline.

`initialize_submodule_config(submodule_config=None, force_kwargs=False, **kwargs)` ¶

Build the config of a submodule from the parent's runtime options and kwargs.

Returns the submodule config instance when its class is known (an instance was passed, or cls_name was given), otherwise the merged kwargs dict for the model class to resolve later.

`init(unet=None, movq=None, *, sample_size=None, batch_size=None, guidance_scale=None, image_size=None, img_height=None, img_width=None, height=None, width=None, **kwargs)` ¶

Parameters:

Name	Type	Description	Default
`unet`	`RBLNUNet2DConditionModelConfig \| None`	Configuration for the UNet model component. Initialized as RBLNUNet2DConditionModelConfig if not provided.	`None`
`movq`	`RBLNVQModelConfig \| None`	Configuration for the MoVQ (VQ-GAN) model component. Initialized as RBLNVQModelConfig if not provided.	`None`
`sample_size`	`tuple[int, int] \| None`	Spatial dimensions for the UNet model.	`None`
`batch_size`	`int \| None`	Batch size for inference, applied to all submodules.	`None`
`guidance_scale`	`float \| None`	Scale for classifier-free guidance.	`None`
`image_size`	`tuple[int, int] \| None`	Dimensions for the generated images. Cannot be used together with img_height/img_width.	`None`
`img_height`	`int \| None`	Height of the generated images.	`None`
`img_width`	`int \| None`	Width of the generated images.	`None`
`height`	`int \| None`	Height of the generated images.	`None`
`width`	`int \| None`	Width of the generated images.	`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.

`get_load_overrides()` ¶

Extract the options that must cross the load boundary, for passing to from_pretrained.

Returns the explicitly-set runtime options (device, device_map, create_runtimes, activate_profiler, timeout) and the subclass-specific non-save attributes (load-behavior flags such as _load_visual_runtime, which are never serialized so they can only travel with the caller), recursively including submodules. Compile-time attributes are not included: when loading a compiled model they come from the artifact's rbln_config.json, and this dict is merged on top of it.

Returns:

Type	Description
`dict[str, Any]`	Dictionary of load overrides, keyed like the `rbln_config` dict accepted by `from_pretrained`.

`from_pretrained(path, rbln_config=None, return_unused_kwargs=False, **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
`rbln_config`	`dict[str, Any] \| None`	Additional configuration to override.	`None`
`return_unused_kwargs`	`bool`	Whether to return unused kwargs.	`False`
`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`	`Self \| tuple[Self, dict[str, Any]]`	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.

Examples:

config = RBLNResNetForImageClassificationConfig.from_pretrained("/path/to/model")

`RBLNKandinskyV22PriorPipelineConfig` ¶

Bases: RBLNModelConfig

Configuration class for the Kandinsky V2.2 Prior pipeline.

`init(text_encoder=None, image_encoder=None, prior=None, *, batch_size=None, guidance_scale=None, **kwargs)` ¶

Initialize a configuration for Kandinsky 2.2 prior pipeline optimized for RBLN NPU.

This configuration sets up the prior components of the Kandinsky 2.2 architecture, which includes text and image encoders along with a prior transformer that maps text/image embeddings to latent representations used to condition the diffusion process.

Parameters:

Name	Type	Description	Default
`text_encoder`	`RBLNCLIPTextModelWithProjectionConfig \| None`	Configuration for the text encoder component. Initialized as RBLNCLIPTextModelWithProjectionConfig if not provided.	`None`
`image_encoder`	`RBLNCLIPVisionModelWithProjectionConfig \| None`	Configuration for the image encoder component. Initialized as RBLNCLIPVisionModelWithProjectionConfig if not provided.	`None`
`prior`	`RBLNPriorTransformerConfig \| None`	Configuration for the prior transformer component. Initialized as RBLNPriorTransformerConfig if not provided.	`None`
`batch_size`	`int \| None`	Batch size for inference, applied to all submodules.	`None`
`guidance_scale`	`float \| None`	Scale for classifier-free guidance.	`None`
`kwargs`	`Any`	Additional arguments passed to the parent RBLNModelConfig.	`{}`

Note

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

`from_pretrained(path, rbln_config=None, return_unused_kwargs=False, **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
`rbln_config`	`dict[str, Any] \| None`	Additional configuration to override.	`None`
`return_unused_kwargs`	`bool`	Whether to return unused kwargs.	`False`
`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`	`Self \| tuple[Self, dict[str, Any]]`	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.

Examples:

config = RBLNResNetForImageClassificationConfig.from_pretrained("/path/to/model")

`initialize_submodule_config(submodule_config=None, force_kwargs=False, **kwargs)` ¶

Build the config of a submodule from the parent's runtime options and kwargs.

Returns the submodule config instance when its class is known (an instance was passed, or cls_name was given), otherwise the merged kwargs dict for the model class to resolve later.

`get_load_overrides()` ¶

Extract the options that must cross the load boundary, for passing to from_pretrained.

Returns the explicitly-set runtime options (device, device_map, create_runtimes, activate_profiler, timeout) and the subclass-specific non-save attributes (load-behavior flags such as _load_visual_runtime, which are never serialized so they can only travel with the caller), recursively including submodules. Compile-time attributes are not included: when loading a compiled model they come from the artifact's rbln_config.json, and this dict is merged on top of it.

Returns:

Type	Description
`dict[str, Any]`	Dictionary of load overrides, keyed like the `rbln_config` dict accepted by `from_pretrained`.

`RBLNKandinskyV22CombinedPipelineBaseConfig` ¶

Bases: RBLNModelConfig

Base configuration class for Kandinsky V2.2 combined pipelines.

`init(prior_pipe=None, decoder_pipe=None, *, sample_size=None, image_size=None, batch_size=None, img_height=None, img_width=None, height=None, width=None, guidance_scale=None, prior_prior=None, prior_image_encoder=None, prior_text_encoder=None, unet=None, movq=None, **kwargs)` ¶

Initialize a configuration for combined Kandinsky 2.2 pipelines optimized for RBLN NPU.

This configuration integrates both the prior and decoder components of Kandinsky 2.2 into a unified pipeline, allowing for end-to-end text-to-image generation in a single model. It combines the text/image encoding, prior mapping, and diffusion steps together.

Parameters:

Name	Type	Description	Default
`prior_pipe`	`RBLNKandinskyV22PriorPipelineConfig \| None`	Configuration for the prior pipeline. Initialized as RBLNKandinskyV22PriorPipelineConfig if not provided.	`None`
`decoder_pipe`	`RBLNKandinskyV22PipelineConfig \| None`	Configuration for the decoder pipeline. Initialized as RBLNKandinskyV22PipelineConfig if not provided.	`None`
`sample_size`	`tuple[int, int] \| None`	Spatial dimensions for the UNet model.	`None`
`image_size`	`tuple[int, int] \| None`	Dimensions for the generated images. Cannot be used together with img_height/img_width.	`None`
`batch_size`	`int \| None`	Batch size for inference, applied to all submodules.	`None`
`img_height`	`int \| None`	Height of the generated images.	`None`
`img_width`	`int \| None`	Width of the generated images.	`None`
`height`	`int \| None`	Height of the generated images.	`None`
`width`	`int \| None`	Width of the generated images.	`None`
`guidance_scale`	`float \| None`	Scale for classifier-free guidance.	`None`
`prior_prior`	`RBLNPriorTransformerConfig \| None`	Direct configuration for the prior transformer. Used if prior_pipe is not provided.	`None`
`prior_image_encoder`	`RBLNCLIPVisionModelWithProjectionConfig \| None`	Direct configuration for the image encoder. Used if prior_pipe is not provided.	`None`
`prior_text_encoder`	`RBLNCLIPTextModelWithProjectionConfig \| None`	Direct configuration for the text encoder. Used if prior_pipe is not provided.	`None`
`unet`	`RBLNUNet2DConditionModelConfig \| None`	Direct configuration for the UNet. Used if decoder_pipe is not provided.	`None`
`movq`	`RBLNVQModelConfig \| None`	Direct configuration for the MoVQ (VQ-GAN) model. Used if decoder_pipe is not provided.	`None`
`kwargs`	`Any`	Additional arguments passed to the parent RBLNModelConfig.	`{}`

`from_pretrained(path, rbln_config=None, return_unused_kwargs=False, **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
`rbln_config`	`dict[str, Any] \| None`	Additional configuration to override.	`None`
`return_unused_kwargs`	`bool`	Whether to return unused kwargs.	`False`
`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`	`Self \| tuple[Self, dict[str, Any]]`	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.

Examples:

config = RBLNResNetForImageClassificationConfig.from_pretrained("/path/to/model")

`initialize_submodule_config(submodule_config=None, force_kwargs=False, **kwargs)` ¶

Build the config of a submodule from the parent's runtime options and kwargs.

Returns the submodule config instance when its class is known (an instance was passed, or cls_name was given), otherwise the merged kwargs dict for the model class to resolve later.

`get_load_overrides()` ¶

Extract the options that must cross the load boundary, for passing to from_pretrained.

Returns the explicitly-set runtime options (device, device_map, create_runtimes, activate_profiler, timeout) and the subclass-specific non-save attributes (load-behavior flags such as _load_visual_runtime, which are never serialized so they can only travel with the caller), recursively including submodules. Compile-time attributes are not included: when loading a compiled model they come from the artifact's rbln_config.json, and this dict is merged on top of it.

Returns:

Type	Description
`dict[str, Any]`	Dictionary of load overrides, keyed like the `rbln_config` dict accepted by `from_pretrained`.

`RBLNKandinskyV22CombinedPipelineConfig` ¶

Bases: RBLNKandinskyV22CombinedPipelineBaseConfig

Configuration class for the Kandinsky V2.2 combined text-to-image pipeline.

`initialize_submodule_config(submodule_config=None, force_kwargs=False, **kwargs)` ¶

Build the config of a submodule from the parent's runtime options and kwargs.

Returns the submodule config instance when its class is known (an instance was passed, or cls_name was given), otherwise the merged kwargs dict for the model class to resolve later.

`init(prior_pipe=None, decoder_pipe=None, *, sample_size=None, image_size=None, batch_size=None, img_height=None, img_width=None, height=None, width=None, guidance_scale=None, prior_prior=None, prior_image_encoder=None, prior_text_encoder=None, unet=None, movq=None, **kwargs)` ¶

Initialize a configuration for combined Kandinsky 2.2 pipelines optimized for RBLN NPU.

This configuration integrates both the prior and decoder components of Kandinsky 2.2 into a unified pipeline, allowing for end-to-end text-to-image generation in a single model. It combines the text/image encoding, prior mapping, and diffusion steps together.

Parameters:

Name	Type	Description	Default
`prior_pipe`	`RBLNKandinskyV22PriorPipelineConfig \| None`	Configuration for the prior pipeline. Initialized as RBLNKandinskyV22PriorPipelineConfig if not provided.	`None`
`decoder_pipe`	`RBLNKandinskyV22PipelineConfig \| None`	Configuration for the decoder pipeline. Initialized as RBLNKandinskyV22PipelineConfig if not provided.	`None`
`sample_size`	`tuple[int, int] \| None`	Spatial dimensions for the UNet model.	`None`
`image_size`	`tuple[int, int] \| None`	Dimensions for the generated images. Cannot be used together with img_height/img_width.	`None`
`batch_size`	`int \| None`	Batch size for inference, applied to all submodules.	`None`
`img_height`	`int \| None`	Height of the generated images.	`None`
`img_width`	`int \| None`	Width of the generated images.	`None`
`height`	`int \| None`	Height of the generated images.	`None`
`width`	`int \| None`	Width of the generated images.	`None`
`guidance_scale`	`float \| None`	Scale for classifier-free guidance.	`None`
`prior_prior`	`RBLNPriorTransformerConfig \| None`	Direct configuration for the prior transformer. Used if prior_pipe is not provided.	`None`
`prior_image_encoder`	`RBLNCLIPVisionModelWithProjectionConfig \| None`	Direct configuration for the image encoder. Used if prior_pipe is not provided.	`None`
`prior_text_encoder`	`RBLNCLIPTextModelWithProjectionConfig \| None`	Direct configuration for the text encoder. Used if prior_pipe is not provided.	`None`
`unet`	`RBLNUNet2DConditionModelConfig \| None`	Direct configuration for the UNet. Used if decoder_pipe is not provided.	`None`
`movq`	`RBLNVQModelConfig \| None`	Direct configuration for the MoVQ (VQ-GAN) model. Used if decoder_pipe is not provided.	`None`
`kwargs`	`Any`	Additional arguments passed to the parent RBLNModelConfig.	`{}`

`get_load_overrides()` ¶

Extract the options that must cross the load boundary, for passing to from_pretrained.

Returns the explicitly-set runtime options (device, device_map, create_runtimes, activate_profiler, timeout) and the subclass-specific non-save attributes (load-behavior flags such as _load_visual_runtime, which are never serialized so they can only travel with the caller), recursively including submodules. Compile-time attributes are not included: when loading a compiled model they come from the artifact's rbln_config.json, and this dict is merged on top of it.

Returns:

Type	Description
`dict[str, Any]`	Dictionary of load overrides, keyed like the `rbln_config` dict accepted by `from_pretrained`.

`from_pretrained(path, rbln_config=None, return_unused_kwargs=False, **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
`rbln_config`	`dict[str, Any] \| None`	Additional configuration to override.	`None`
`return_unused_kwargs`	`bool`	Whether to return unused kwargs.	`False`
`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`	`Self \| tuple[Self, dict[str, Any]]`	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.

Examples:

config = RBLNResNetForImageClassificationConfig.from_pretrained("/path/to/model")

`RBLNKandinskyV22InpaintCombinedPipelineConfig` ¶

Bases: RBLNKandinskyV22CombinedPipelineBaseConfig

Configuration class for the Kandinsky V2.2 combined inpainting pipeline.

`initialize_submodule_config(submodule_config=None, force_kwargs=False, **kwargs)` ¶

Build the config of a submodule from the parent's runtime options and kwargs.

Returns the submodule config instance when its class is known (an instance was passed, or cls_name was given), otherwise the merged kwargs dict for the model class to resolve later.

`init(prior_pipe=None, decoder_pipe=None, *, sample_size=None, image_size=None, batch_size=None, img_height=None, img_width=None, height=None, width=None, guidance_scale=None, prior_prior=None, prior_image_encoder=None, prior_text_encoder=None, unet=None, movq=None, **kwargs)` ¶

Initialize a configuration for combined Kandinsky 2.2 pipelines optimized for RBLN NPU.

This configuration integrates both the prior and decoder components of Kandinsky 2.2 into a unified pipeline, allowing for end-to-end text-to-image generation in a single model. It combines the text/image encoding, prior mapping, and diffusion steps together.

Parameters:

Name	Type	Description	Default
`prior_pipe`	`RBLNKandinskyV22PriorPipelineConfig \| None`	Configuration for the prior pipeline. Initialized as RBLNKandinskyV22PriorPipelineConfig if not provided.	`None`
`decoder_pipe`	`RBLNKandinskyV22PipelineConfig \| None`	Configuration for the decoder pipeline. Initialized as RBLNKandinskyV22PipelineConfig if not provided.	`None`
`sample_size`	`tuple[int, int] \| None`	Spatial dimensions for the UNet model.	`None`
`image_size`	`tuple[int, int] \| None`	Dimensions for the generated images. Cannot be used together with img_height/img_width.	`None`
`batch_size`	`int \| None`	Batch size for inference, applied to all submodules.	`None`
`img_height`	`int \| None`	Height of the generated images.	`None`
`img_width`	`int \| None`	Width of the generated images.	`None`
`height`	`int \| None`	Height of the generated images.	`None`
`width`	`int \| None`	Width of the generated images.	`None`
`guidance_scale`	`float \| None`	Scale for classifier-free guidance.	`None`
`prior_prior`	`RBLNPriorTransformerConfig \| None`	Direct configuration for the prior transformer. Used if prior_pipe is not provided.	`None`
`prior_image_encoder`	`RBLNCLIPVisionModelWithProjectionConfig \| None`	Direct configuration for the image encoder. Used if prior_pipe is not provided.	`None`
`prior_text_encoder`	`RBLNCLIPTextModelWithProjectionConfig \| None`	Direct configuration for the text encoder. Used if prior_pipe is not provided.	`None`
`unet`	`RBLNUNet2DConditionModelConfig \| None`	Direct configuration for the UNet. Used if decoder_pipe is not provided.	`None`
`movq`	`RBLNVQModelConfig \| None`	Direct configuration for the MoVQ (VQ-GAN) model. Used if decoder_pipe is not provided.	`None`
`kwargs`	`Any`	Additional arguments passed to the parent RBLNModelConfig.	`{}`

`get_load_overrides()` ¶

Extract the options that must cross the load boundary, for passing to from_pretrained.

Returns the explicitly-set runtime options (device, device_map, create_runtimes, activate_profiler, timeout) and the subclass-specific non-save attributes (load-behavior flags such as _load_visual_runtime, which are never serialized so they can only travel with the caller), recursively including submodules. Compile-time attributes are not included: when loading a compiled model they come from the artifact's rbln_config.json, and this dict is merged on top of it.

Returns:

Type	Description
`dict[str, Any]`	Dictionary of load overrides, keyed like the `rbln_config` dict accepted by `from_pretrained`.

`from_pretrained(path, rbln_config=None, return_unused_kwargs=False, **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
`rbln_config`	`dict[str, Any] \| None`	Additional configuration to override.	`None`
`return_unused_kwargs`	`bool`	Whether to return unused kwargs.	`False`
`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`	`Self \| tuple[Self, dict[str, Any]]`	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.

Examples:

config = RBLNResNetForImageClassificationConfig.from_pretrained("/path/to/model")

`RBLNKandinskyV22Img2ImgCombinedPipelineConfig` ¶

Bases: RBLNKandinskyV22CombinedPipelineBaseConfig

Configuration class for the Kandinsky V2.2 combined image-to-image pipeline.

`initialize_submodule_config(submodule_config=None, force_kwargs=False, **kwargs)` ¶

Build the config of a submodule from the parent's runtime options and kwargs.

Returns the submodule config instance when its class is known (an instance was passed, or cls_name was given), otherwise the merged kwargs dict for the model class to resolve later.

`init(prior_pipe=None, decoder_pipe=None, *, sample_size=None, image_size=None, batch_size=None, img_height=None, img_width=None, height=None, width=None, guidance_scale=None, prior_prior=None, prior_image_encoder=None, prior_text_encoder=None, unet=None, movq=None, **kwargs)` ¶

Initialize a configuration for combined Kandinsky 2.2 pipelines optimized for RBLN NPU.

This configuration integrates both the prior and decoder components of Kandinsky 2.2 into a unified pipeline, allowing for end-to-end text-to-image generation in a single model. It combines the text/image encoding, prior mapping, and diffusion steps together.

Parameters:

Name	Type	Description	Default
`prior_pipe`	`RBLNKandinskyV22PriorPipelineConfig \| None`	Configuration for the prior pipeline. Initialized as RBLNKandinskyV22PriorPipelineConfig if not provided.	`None`
`decoder_pipe`	`RBLNKandinskyV22PipelineConfig \| None`	Configuration for the decoder pipeline. Initialized as RBLNKandinskyV22PipelineConfig if not provided.	`None`
`sample_size`	`tuple[int, int] \| None`	Spatial dimensions for the UNet model.	`None`
`image_size`	`tuple[int, int] \| None`	Dimensions for the generated images. Cannot be used together with img_height/img_width.	`None`
`batch_size`	`int \| None`	Batch size for inference, applied to all submodules.	`None`
`img_height`	`int \| None`	Height of the generated images.	`None`
`img_width`	`int \| None`	Width of the generated images.	`None`
`height`	`int \| None`	Height of the generated images.	`None`
`width`	`int \| None`	Width of the generated images.	`None`
`guidance_scale`	`float \| None`	Scale for classifier-free guidance.	`None`
`prior_prior`	`RBLNPriorTransformerConfig \| None`	Direct configuration for the prior transformer. Used if prior_pipe is not provided.	`None`
`prior_image_encoder`	`RBLNCLIPVisionModelWithProjectionConfig \| None`	Direct configuration for the image encoder. Used if prior_pipe is not provided.	`None`
`prior_text_encoder`	`RBLNCLIPTextModelWithProjectionConfig \| None`	Direct configuration for the text encoder. Used if prior_pipe is not provided.	`None`
`unet`	`RBLNUNet2DConditionModelConfig \| None`	Direct configuration for the UNet. Used if decoder_pipe is not provided.	`None`
`movq`	`RBLNVQModelConfig \| None`	Direct configuration for the MoVQ (VQ-GAN) model. Used if decoder_pipe is not provided.	`None`
`kwargs`	`Any`	Additional arguments passed to the parent RBLNModelConfig.	`{}`

`get_load_overrides()` ¶

Extract the options that must cross the load boundary, for passing to from_pretrained.

Returns the explicitly-set runtime options (device, device_map, create_runtimes, activate_profiler, timeout) and the subclass-specific non-save attributes (load-behavior flags such as _load_visual_runtime, which are never serialized so they can only travel with the caller), recursively including submodules. Compile-time attributes are not included: when loading a compiled model they come from the artifact's rbln_config.json, and this dict is merged on top of it.

Returns:

Type	Description
`dict[str, Any]`	Dictionary of load overrides, keyed like the `rbln_config` dict accepted by `from_pretrained`.

`from_pretrained(path, rbln_config=None, return_unused_kwargs=False, **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
`rbln_config`	`dict[str, Any] \| None`	Additional configuration to override.	`None`
`return_unused_kwargs`	`bool`	Whether to return unused kwargs.	`False`
`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`	`Self \| tuple[Self, dict[str, Any]]`	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.

Examples:

config = RBLNResNetForImageClassificationConfig.from_pretrained("/path/to/model")

Kandinsky V2.2¶

지원하는 파이프라인¶

중요: guidance scale을 위한 배치 크기 설정¶

기본 동작¶

예제: 명시적으로 배치 크기 설정하기(guidance scale = 1.0)¶

사용 예제¶

방법 1: 개별 Prior 및 Decoder 파이프라인 사용¶

방법 2: 결합 파이프라인 사용¶

API 참조¶

RBLNKandinskyV22Pipeline ¶

from_pretrained(model_id, *, export=None, model_save_dir=None, rbln_config=None, lora_ids=None, lora_weights_names=None, lora_scales=None, **kwargs) classmethod ¶

RBLNKandinskyV22PriorPipeline ¶

from_pretrained(model_id, *, export=None, model_save_dir=None, rbln_config=None, lora_ids=None, lora_weights_names=None, lora_scales=None, **kwargs) classmethod ¶

RBLNKandinskyV22Img2ImgPipeline ¶

from_pretrained(model_id, *, export=None, model_save_dir=None, rbln_config=None, lora_ids=None, lora_weights_names=None, lora_scales=None, **kwargs) classmethod ¶

RBLNKandinskyV22InpaintPipeline ¶

from_pretrained(model_id, *, export=None, model_save_dir=None, rbln_config=None, lora_ids=None, lora_weights_names=None, lora_scales=None, **kwargs) classmethod ¶

RBLNKandinskyV22CombinedPipeline ¶

from_pretrained(model_id, *, export=None, model_save_dir=None, rbln_config=None, lora_ids=None, lora_weights_names=None, lora_scales=None, **kwargs) classmethod ¶

RBLNKandinskyV22Img2ImgCombinedPipeline ¶

from_pretrained(model_id, *, export=None, model_save_dir=None, rbln_config=None, lora_ids=None, lora_weights_names=None, lora_scales=None, **kwargs) classmethod ¶

RBLNKandinskyV22InpaintCombinedPipeline ¶

from_pretrained(model_id, *, export=None, model_save_dir=None, rbln_config=None, lora_ids=None, lora_weights_names=None, lora_scales=None, **kwargs) classmethod ¶

RBLNKandinskyV22PipelineBaseConfig ¶

__init__(unet=None, movq=None, *, sample_size=None, batch_size=None, guidance_scale=None, image_size=None, img_height=None, img_width=None, height=None, width=None, **kwargs) ¶

from_pretrained(path, rbln_config=None, return_unused_kwargs=False, **kwargs) classmethod ¶

initialize_submodule_config(submodule_config=None, force_kwargs=False, **kwargs) ¶

get_load_overrides() ¶

RBLNKandinskyV22PipelineConfig ¶

initialize_submodule_config(submodule_config=None, force_kwargs=False, **kwargs) ¶

__init__(unet=None, movq=None, *, sample_size=None, batch_size=None, guidance_scale=None, image_size=None, img_height=None, img_width=None, height=None, width=None, **kwargs) ¶

get_load_overrides() ¶

from_pretrained(path, rbln_config=None, return_unused_kwargs=False, **kwargs) classmethod ¶

RBLNKandinskyV22Img2ImgPipelineConfig ¶

initialize_submodule_config(submodule_config=None, force_kwargs=False, **kwargs) ¶

__init__(unet=None, movq=None, *, sample_size=None, batch_size=None, guidance_scale=None, image_size=None, img_height=None, img_width=None, height=None, width=None, **kwargs) ¶

get_load_overrides() ¶

from_pretrained(path, rbln_config=None, return_unused_kwargs=False, **kwargs) classmethod ¶

RBLNKandinskyV22InpaintPipelineConfig ¶

initialize_submodule_config(submodule_config=None, force_kwargs=False, **kwargs) ¶

__init__(unet=None, movq=None, *, sample_size=None, batch_size=None, guidance_scale=None, image_size=None, img_height=None, img_width=None, height=None, width=None, **kwargs) ¶

get_load_overrides() ¶

from_pretrained(path, rbln_config=None, return_unused_kwargs=False, **kwargs) classmethod ¶

RBLNKandinskyV22PriorPipelineConfig ¶

__init__(text_encoder=None, image_encoder=None, prior=None, *, batch_size=None, guidance_scale=None, **kwargs) ¶

from_pretrained(path, rbln_config=None, return_unused_kwargs=False, **kwargs) classmethod ¶

initialize_submodule_config(submodule_config=None, force_kwargs=False, **kwargs) ¶

get_load_overrides() ¶

RBLNKandinskyV22CombinedPipelineBaseConfig ¶

__init__(prior_pipe=None, decoder_pipe=None, *, sample_size=None, image_size=None, batch_size=None, img_height=None, img_width=None, height=None, width=None, guidance_scale=None, prior_prior=None, prior_image_encoder=None, prior_text_encoder=None, unet=None, movq=None, **kwargs) ¶

from_pretrained(path, rbln_config=None, return_unused_kwargs=False, **kwargs) classmethod ¶

initialize_submodule_config(submodule_config=None, force_kwargs=False, **kwargs) ¶

get_load_overrides() ¶

RBLNKandinskyV22CombinedPipelineConfig ¶

initialize_submodule_config(submodule_config=None, force_kwargs=False, **kwargs) ¶

__init__(prior_pipe=None, decoder_pipe=None, *, sample_size=None, image_size=None, batch_size=None, img_height=None, img_width=None, height=None, width=None, guidance_scale=None, prior_prior=None, prior_image_encoder=None, prior_text_encoder=None, unet=None, movq=None, **kwargs) ¶

get_load_overrides() ¶

from_pretrained(path, rbln_config=None, return_unused_kwargs=False, **kwargs) classmethod ¶

RBLNKandinskyV22InpaintCombinedPipelineConfig ¶

initialize_submodule_config(submodule_config=None, force_kwargs=False, **kwargs) ¶

__init__(prior_pipe=None, decoder_pipe=None, *, sample_size=None, image_size=None, batch_size=None, img_height=None, img_width=None, height=None, width=None, guidance_scale=None, prior_prior=None, prior_image_encoder=None, prior_text_encoder=None, unet=None, movq=None, **kwargs) ¶

get_load_overrides() ¶

from_pretrained(path, rbln_config=None, return_unused_kwargs=False, **kwargs) classmethod ¶

RBLNKandinskyV22Img2ImgCombinedPipelineConfig ¶

initialize_submodule_config(submodule_config=None, force_kwargs=False, **kwargs) ¶

__init__(prior_pipe=None, decoder_pipe=None, *, sample_size=None, image_size=None, batch_size=None, img_height=None, img_width=None, height=None, width=None, guidance_scale=None, prior_prior=None, prior_image_encoder=None, prior_text_encoder=None, unet=None, movq=None, **kwargs) ¶

get_load_overrides() ¶

from_pretrained(path, rbln_config=None, return_unused_kwargs=False, **kwargs) classmethod ¶

`RBLNKandinskyV22Pipeline` ¶

`from_pretrained(model_id, *, export=None, model_save_dir=None, rbln_config=None, lora_ids=None, lora_weights_names=None, lora_scales=None, **kwargs)` `classmethod` ¶

`RBLNKandinskyV22PriorPipeline` ¶

`from_pretrained(model_id, *, export=None, model_save_dir=None, rbln_config=None, lora_ids=None, lora_weights_names=None, lora_scales=None, **kwargs)` `classmethod` ¶

`RBLNKandinskyV22Img2ImgPipeline` ¶

`from_pretrained(model_id, *, export=None, model_save_dir=None, rbln_config=None, lora_ids=None, lora_weights_names=None, lora_scales=None, **kwargs)` `classmethod` ¶

`RBLNKandinskyV22InpaintPipeline` ¶

`from_pretrained(model_id, *, export=None, model_save_dir=None, rbln_config=None, lora_ids=None, lora_weights_names=None, lora_scales=None, **kwargs)` `classmethod` ¶

`RBLNKandinskyV22CombinedPipeline` ¶

`from_pretrained(model_id, *, export=None, model_save_dir=None, rbln_config=None, lora_ids=None, lora_weights_names=None, lora_scales=None, **kwargs)` `classmethod` ¶

`RBLNKandinskyV22Img2ImgCombinedPipeline` ¶

`from_pretrained(model_id, *, export=None, model_save_dir=None, rbln_config=None, lora_ids=None, lora_weights_names=None, lora_scales=None, **kwargs)` `classmethod` ¶

`RBLNKandinskyV22InpaintCombinedPipeline` ¶

`from_pretrained(model_id, *, export=None, model_save_dir=None, rbln_config=None, lora_ids=None, lora_weights_names=None, lora_scales=None, **kwargs)` `classmethod` ¶

`RBLNKandinskyV22PipelineBaseConfig` ¶

`init(unet=None, movq=None, *, sample_size=None, batch_size=None, guidance_scale=None, image_size=None, img_height=None, img_width=None, height=None, width=None, **kwargs)` ¶

`from_pretrained(path, rbln_config=None, return_unused_kwargs=False, **kwargs)` `classmethod` ¶

`initialize_submodule_config(submodule_config=None, force_kwargs=False, **kwargs)` ¶

`get_load_overrides()` ¶

`RBLNKandinskyV22PipelineConfig` ¶

`initialize_submodule_config(submodule_config=None, force_kwargs=False, **kwargs)` ¶

`init(unet=None, movq=None, *, sample_size=None, batch_size=None, guidance_scale=None, image_size=None, img_height=None, img_width=None, height=None, width=None, **kwargs)` ¶

`get_load_overrides()` ¶

`from_pretrained(path, rbln_config=None, return_unused_kwargs=False, **kwargs)` `classmethod` ¶

`RBLNKandinskyV22Img2ImgPipelineConfig` ¶

`initialize_submodule_config(submodule_config=None, force_kwargs=False, **kwargs)` ¶

`init(unet=None, movq=None, *, sample_size=None, batch_size=None, guidance_scale=None, image_size=None, img_height=None, img_width=None, height=None, width=None, **kwargs)` ¶

`get_load_overrides()` ¶

`from_pretrained(path, rbln_config=None, return_unused_kwargs=False, **kwargs)` `classmethod` ¶

`RBLNKandinskyV22InpaintPipelineConfig` ¶

`initialize_submodule_config(submodule_config=None, force_kwargs=False, **kwargs)` ¶

`init(unet=None, movq=None, *, sample_size=None, batch_size=None, guidance_scale=None, image_size=None, img_height=None, img_width=None, height=None, width=None, **kwargs)` ¶

`get_load_overrides()` ¶

`from_pretrained(path, rbln_config=None, return_unused_kwargs=False, **kwargs)` `classmethod` ¶

`RBLNKandinskyV22PriorPipelineConfig` ¶

`init(text_encoder=None, image_encoder=None, prior=None, *, batch_size=None, guidance_scale=None, **kwargs)` ¶

`from_pretrained(path, rbln_config=None, return_unused_kwargs=False, **kwargs)` `classmethod` ¶

`initialize_submodule_config(submodule_config=None, force_kwargs=False, **kwargs)` ¶

`get_load_overrides()` ¶

`RBLNKandinskyV22CombinedPipelineBaseConfig` ¶

`init(prior_pipe=None, decoder_pipe=None, *, sample_size=None, image_size=None, batch_size=None, img_height=None, img_width=None, height=None, width=None, guidance_scale=None, prior_prior=None, prior_image_encoder=None, prior_text_encoder=None, unet=None, movq=None, **kwargs)` ¶

`from_pretrained(path, rbln_config=None, return_unused_kwargs=False, **kwargs)` `classmethod` ¶

`initialize_submodule_config(submodule_config=None, force_kwargs=False, **kwargs)` ¶

`get_load_overrides()` ¶

`RBLNKandinskyV22CombinedPipelineConfig` ¶

`initialize_submodule_config(submodule_config=None, force_kwargs=False, **kwargs)` ¶

`init(prior_pipe=None, decoder_pipe=None, *, sample_size=None, image_size=None, batch_size=None, img_height=None, img_width=None, height=None, width=None, guidance_scale=None, prior_prior=None, prior_image_encoder=None, prior_text_encoder=None, unet=None, movq=None, **kwargs)` ¶

`get_load_overrides()` ¶

`from_pretrained(path, rbln_config=None, return_unused_kwargs=False, **kwargs)` `classmethod` ¶

`RBLNKandinskyV22InpaintCombinedPipelineConfig` ¶

`initialize_submodule_config(submodule_config=None, force_kwargs=False, **kwargs)` ¶

`init(prior_pipe=None, decoder_pipe=None, *, sample_size=None, image_size=None, batch_size=None, img_height=None, img_width=None, height=None, width=None, guidance_scale=None, prior_prior=None, prior_image_encoder=None, prior_text_encoder=None, unet=None, movq=None, **kwargs)` ¶

`get_load_overrides()` ¶

`from_pretrained(path, rbln_config=None, return_unused_kwargs=False, **kwargs)` `classmethod` ¶

`RBLNKandinskyV22Img2ImgCombinedPipelineConfig` ¶

`initialize_submodule_config(submodule_config=None, force_kwargs=False, **kwargs)` ¶

`init(prior_pipe=None, decoder_pipe=None, *, sample_size=None, image_size=None, batch_size=None, img_height=None, img_width=None, height=None, width=None, guidance_scale=None, prior_prior=None, prior_image_encoder=None, prior_text_encoder=None, unet=None, movq=None, **kwargs)` ¶

`get_load_overrides()` ¶

`from_pretrained(path, rbln_config=None, return_unused_kwargs=False, **kwargs)` `classmethod` ¶