Comfy-Org/ComfyUI · error · ValueError

The 'end_frame' input cannot be used simultaneously with 're

Error message

The 'end_frame' input cannot be used simultaneously with 'reference_images'.

What it means

Raised by OmniProFirstLastFrameNode.execute when both end_frame and reference_images are connected. The Kling omni API treats end_frame as a special typed image entry and reference_images as generic reference entries; sending both in one request is not a supported combination, so the node rejects it before uploading anything.

Source

Thrown at comfy_api_nodes/nodes_kling.py:880

        storyboards: dict | None = None,
        generate_audio: bool = False,
        seed: int = 0,
    ) -> IO.NodeOutput:
        _ = seed
        if model_name == "kling-video-o1":
            if duration > 10:
                raise ValueError("kling-video-o1 does not support durations greater than 10 seconds.")
            if generate_audio:
                raise ValueError("kling-video-o1 does not support audio generation.")
            if resolution == "4k":
                raise ValueError("kling-video-o1 does not support 4k resolution.")
        stories_enabled = storyboards is not None and storyboards["storyboards"] != "disabled"
        if stories_enabled and model_name == "kling-video-o1":
            raise ValueError("kling-video-o1 does not support storyboards.")
        prompt = normalize_omni_prompt_references(prompt)
        validate_string(prompt, strip_whitespace=True, min_length=0 if stories_enabled else 1, max_length=2500)
        if end_frame is not None and reference_images is not None:
            raise ValueError("The 'end_frame' input cannot be used simultaneously with 'reference_images'.")
        if end_frame is not None and stories_enabled:
            raise ValueError("The 'end_frame' input cannot be used simultaneously with storyboards.")
        if (
            model_name == "kling-video-o1"
            and duration not in (5, 10)
            and end_frame is None
            and reference_images is None
        ):
            raise ValueError(
                "Duration is only supported for 5 or 10 seconds if there is no end frame or reference images."
            )

        multi_shot = None
        multi_prompt_list = None
        if stories_enabled:
            count = int(storyboards["storyboards"].split()[0])
            multi_shot = True
            multi_prompt_list = []

View on GitHub (pinned to 1c6d8d45b3)

Solutions

1. Disconnect one of the two inputs: keep end_frame for last-frame control, or keep reference_images for style/subject references
2. If you need both a reference and a specific final frame, bake the reference into first_frame or the prompt and keep only end_frame
3. Run two passes: generate with reference_images, then use the last output frame as end_frame in a second node

Example fix

// before
end_frame = load_last_frame.output
reference_images = load_reference_batch.output

// after
end_frame = load_last_frame.output
reference_images = None  # pick one control mode per request

def end_frame_reference_ok(end_frame, reference_images) -> bool:
    return end_frame is None or reference_images is None

When it happens

Trigger: Connecting tensors to both the end_frame input and the reference_images input of the OmniPro first/last frame node in the same graph.

Common situations: A template workflow uses reference_images for character consistency; a user adds an end frame for a controlled ending without disconnecting the reference image batch, or vice versa after copying a node.

Related errors

• The 'end_frame' input cannot be used simultaneously with sto
• Duration is only supported for 5 or 10 seconds if there is n
• The maximum number of reference images allowed is 6.
• The maximum number of reference images is 7.
• The maximum number of reference images allowed with a video

AI-assisted analysis of Comfy-Org/ComfyUI@1c6d8d45b3 (2026-08-14). Data as JSON: /api/errors/8b5ad1a5af1bc2b3. Report an issue: GitHub.