Comfy-Org/ComfyUI · warning · ValueError

Currently only one input image is supported.

Error message

Currently only one input image is supported.

What it means

Client-side ValueError in the v1 LTX image-to-video node: the node supports exactly one input image, and get_number_of_images(image) returned something other than 1 (a batch or zero images after mask/crop upstream).

Source

Thrown at comfy_api_nodes/nodes_ltxv.py:362

    @classmethod
    async def execute(
        cls,
        image: Input.Image,
        model: str,
        prompt: str,
        duration: int,
        resolution: str,
        fps: int = 25,
        generate_audio: bool = False,
    ) -> IO.NodeOutput:
        validate_string(prompt, min_length=1, max_length=10000)
        if duration > 10 and (model != "LTX-2 (Fast)" or resolution != "1920x1080" or fps != 25):
            raise ValueError(
                "Durations over 10s are only available for the Fast model at 1920x1080 resolution and 25 FPS."
            )
        if get_number_of_images(image) != 1:
            raise ValueError("Currently only one input image is supported.")
        response = await sync_op_raw(
            cls,
            ApiEndpoint("/proxy/ltx/v1/image-to-video", "POST"),
            data=ExecuteTaskRequest(
                image_uri=(await upload_images_to_comfyapi(cls, image, max_images=1, mime_type="image/png"))[0],
                prompt=prompt,
                model=MODELS_MAP[model],
                duration=duration,
                resolution=resolution,
                fps=fps,
                generate_audio=generate_audio,
            ),
            as_binary=True,
            max_retries=1,
        )
        return IO.NodeOutput(InputImpl.VideoFromFile(BytesIO(response)))

View on GitHub (pinned to 1c6d8d45b3)

Solutions

1. Ensure exactly one image reaches the node: set batch size to 1 in the loader.
2. Insert a Split Image or Image Batch index selection to pick a single frame from a batch.
3. Check upstream nodes for empty outputs (0 images) when the error appears with blank inputs.

Example fix

// before: batched input
image = load_image_batch(path, batch=4)  # shape[0]=4
// after
image = load_image(path)  # shape[0]=1, single image

assert image.shape[0] == 1, f"Expected 1 image, got batch of {image.shape[0]}"

def is_single_image(image: torch.Tensor) -> bool:
    return image.ndim >= 1 and image.shape[0] == 1

When it happens

Trigger: Feeding a batched IMAGE tensor (shape[0] > 1) or an empty tensor into the v1 image-to-video node; common when an upstream node produces batches (e.g. Load Image with batch, empty Latent/crop edge cases).

Common situations: LoadImage batch size > 1; an upstream crop/resize node emitting an empty batch for degenerate inputs; chaining from a node that splits or packs images unexpectedly.

Related errors

• Currently only one last frame image is supported.
• Durations over 10s require a 720p or 1080p resolution and 24
• Durations over 10s are only available for the Fast model at
• Audio duration must be between 2 and 20 seconds, got {audio_
• Total storyboard duration ({duration}s) must be between 3 an

AI-assisted analysis of Comfy-Org/ComfyUI@1c6d8d45b3 (2026-08-14). Data as JSON: /api/errors/81bf5975760e2791. Report an issue: GitHub.