Skip to content

vllm_omni.diffusion.models.longcat_image.pipeline_longcat_image

logger module-attribute 

logger = init_logger(__name__)

LongCatImagePipeline

Bases: Module, CFGParallelMixin, DiffusionPipelineProfilerMixin

default_sample_size instance-attribute 

default_sample_size = 128

device instance-attribute 

device = get_local_device()

do_classifier_free_guidance property 

do_classifier_free_guidance

od_config instance-attribute 

od_config = od_config

prompt_template_encode_prefix instance-attribute 

prompt_template_encode_prefix = "<|im_start|>system\nAs an image captioning expert, generate a descriptive text prompt based on an image content, suitable for input to a text-to-image model.<|im_end|>\n<|im_start|>user\n"

prompt_template_encode_suffix instance-attribute 

prompt_template_encode_suffix = (
    "<|im_end|>\n<|im_start|>assistant\n"
)

scheduler instance-attribute 

scheduler = from_pretrained(
    model,
    subfolder="scheduler",
    local_files_only=local_files_only,
)

text_encoder instance-attribute 

text_encoder = from_pretrained_with_prefetch(
    from_pretrained,
    model,
    subfolder="text_encoder",
    prefetch_list=longcat_subfolders,
    local_files_only=local_files_only,
)

text_processor instance-attribute 

text_processor = from_pretrained(
    model,
    subfolder="tokenizer",
    local_files_only=local_files_only,
)

tokenizer instance-attribute 

tokenizer = from_pretrained(
    model,
    subfolder="tokenizer",
    local_files_only=local_files_only,
)

tokenizer_max_length instance-attribute 

tokenizer_max_length = 512

transformer instance-attribute 

transformer = LongCatImageTransformer2DModel(
    od_config=od_config
)

vae instance-attribute 

vae = to(device)

vae_scale_factor instance-attribute 

vae_scale_factor = (
    2 ** (len(block_out_channels) - 1)
    if getattr(self, "vae", None)
    else 8
)

weights_sources instance-attribute 

weights_sources = [
    ComponentSource(
        model_or_path=model,
        subfolder="transformer",
        revision=None,
        prefix="transformer.",
        fall_back_to_pt=True,
    )
]

cfg_normalize_function 

cfg_normalize_function(
    noise_pred, comb_pred, cfg_renorm_min=0.0
)

Normalize the combined noise prediction.

check_inputs 

check_inputs(
    prompt,
    height,
    width,
    negative_prompt=None,
    prompt_embeds=None,
    negative_prompt_embeds=None,
)

encode_prompt 

encode_prompt(
    prompt: str | list[str] | None = None,
    num_images_per_prompt: int | None = 1,
    prompt_embeds: Tensor | None = None,
) -> tuple[Tensor, Tensor]

forward 

forward(
    req: OmniDiffusionRequest,
    prompt: str | list[str] | None = None,
    negative_prompt: str | list[str] | None = None,
    height: int | None = None,
    width: int | None = None,
    num_inference_steps: int = 50,
    sigmas: list[float] | None = None,
    guidance_scale: float = 4.5,
    num_images_per_prompt: int = 1,
    generator: Generator | list[Generator] | None = None,
    latents: FloatTensor | None = None,
    prompt_embeds: Tensor | None = None,
    negative_prompt_embeds: Tensor | None = None,
    output_type: str | None = "pil",
    return_dict: bool = True,
    joint_attention_kwargs: dict[str, Any] | None = None,
    enable_cfg_renorm: bool | None = True,
    cfg_renorm_min: float | None = 0.0,
    enable_prompt_rewrite: bool | None = True,
) -> DiffusionOutput

load_weights 

load_weights(
    weights: Iterable[tuple[str, Tensor]],
) -> set[str]

Load weights using AutoWeightsLoader for vLLM integration.

prepare_latents 

prepare_latents(
    batch_size,
    num_channels_latents,
    height,
    width,
    dtype,
    device,
    generator,
    latents=None,
)

rewire_prompt 

rewire_prompt(prompt, device)

calculate_shift 

calculate_shift(
    image_seq_len,
    base_seq_len: int = 256,
    max_seq_len: int = 4096,
    base_shift: float = 0.5,
    max_shift: float = 1.15,
)

get_longcat_image_post_process_func 

get_longcat_image_post_process_func(
    od_config: OmniDiffusionConfig,
)

get_prompt_language 

get_prompt_language(prompt)

prepare_pos_ids 

prepare_pos_ids(
    modality_id=0,
    type="text",
    start=(0, 0),
    num_token=None,
    height=None,
    width=None,
) -> Tensor

retrieve_timesteps 

retrieve_timesteps(
    scheduler: SchedulerMixin,
    num_inference_steps: int | None = None,
    device: str | device | None = None,
    timesteps: list[int] | None = None,
    sigmas: list[float] | None = None,
    **kwargs,
) -> tuple[Tensor, int]

Calls the scheduler's set_timesteps method and retrieves timesteps from the scheduler after the call. Handles custom timesteps. Any kwargs will be supplied to scheduler.set_timesteps.

Parameters:

Name Type Description Default
scheduler `SchedulerMixin`

The scheduler to get timesteps from.

 required
num_inference_steps `int`

The number of diffusion steps used when generating samples with a pre-trained model. If used, timesteps must be None.

 None
device `str` or `torch.device`, *optional*

The device to which the timesteps should be moved to. If None, the timesteps are not moved.

 None
timesteps `list[int]`, *optional*

Custom timesteps used to override the timestep spacing strategy of the scheduler. If timesteps is passed, num_inference_steps and sigmas must be None.

 None
sigmas `list[float]`, *optional*

Custom sigmas used to override the timestep spacing strategy of the scheduler. If sigmas is passed, num_inference_steps and timesteps must be None.

 None

Returns:

Type Description
Tensor

tuple[torch.Tensor, int]: A tuple where the first element is the timestep schedule from the scheduler and the
int

second element is the number of inference steps.

split_quotation 

split_quotation(prompt, quote_pairs=None)

Implement a regex-based string splitting algorithm that identifies delimiters defined by single or double quote pairs.

Examples:: >>> prompt_en = "Please write 'Hello' on the blackboard for me." >>> print(split_quotation(prompt_en)) >>> # output: [('Please write ', False), ("'Hello'", True), (' on the blackboard for me.', False)]