Pika 2.2 Image to Video - ComfyUI Native Node Documentation - ComfyUI

ComfyUI Native Pika 2.2 Image to Video Node

The Pika 2.2 Image to Video node connects to Pika’s latest 2.2 API to transform static images into dynamic videos. It preserves the visual features of the original image while adding natural motion based on text prompts.

Parameters

Required Parameters

Parameter	Type	Default	Description
image	Image	-	Input image to convert to video
prompt_text	String	""	Text prompt describing video motion and content
negative_prompt	String	""	Elements to avoid in the video
seed	Integer	0	Random seed for generation
resolution	Select	”1080p”	Output video resolution
duration	Select	”5s”	Length of generated video

Output

Output	Type	Description
VIDEO	Video	Generated video

How It Works

The node sends the input image and parameters (prompts, resolution, duration, etc.) to Pika’s API server as multipart form data. The API processes this and returns the generated video. Users can control the output by adjusting the prompts, negative prompts, random seed and other parameters.

Source Code

[Node Source Code (Updated 2025-05-05)]

class PikaImageToVideoV2_2(PikaNodeBase):
    """Pika 2.2 Image to Video Node."""

    @classmethod
    def INPUT_TYPES(cls):
        return {
            "required": {
                "image": (
                    IO.IMAGE,
                    {"tooltip": "The image to convert to video"},
                ),
                **cls.get_base_inputs_types(PikaBodyGenerate22I2vGenerate22I2vPost),
            },
            "hidden": {
                "auth_token": "AUTH_TOKEN_COMFY_ORG",
            },
        }

    DESCRIPTION = "Sends an image and prompt to the Pika API v2.2 to generate a video."
    RETURN_TYPES = ("VIDEO",)

    def api_call(
        self,
        image: torch.Tensor,
        prompt_text: str,
        negative_prompt: str,
        seed: int,
        resolution: str,
        duration: int,
        auth_token: Optional[str] = None,
    ) -> tuple[VideoFromFile]:
        """API call for Pika 2.2 Image to Video."""
        # Convert image to BytesIO
        image_bytes_io = tensor_to_bytesio(image)
        image_bytes_io.seek(0)  # Reset stream position

        # Prepare file data for multipart upload
        pika_files = {"image": ("image.png", image_bytes_io, "image/png")}

        # Prepare non-file data using the Pydantic model
        pika_request_data = PikaBodyGenerate22I2vGenerate22I2vPost(
            promptText=prompt_text,
            negativePrompt=negative_prompt,
            seed=seed,
            resolution=resolution,
            duration=duration,
        )

        initial_operation = SynchronousOperation(
            endpoint=ApiEndpoint(
                path=PATH_IMAGE_TO_VIDEO,
                method=HttpMethod.POST,
                request_model=PikaBodyGenerate22I2vGenerate22I2vPost,
                response_model=PikaGenerateResponse,
            ),
            request=pika_request_data,
            files=pika_files,
            content_type="multipart/form-data",
            auth_token=auth_token,
        )

        return self.execute_task(initial_operation, auth_token)