sgl-project/sglang · error · ValueError

keyframe payload requires an integer frame_count

Error message

keyframe payload requires an integer frame_count

What it means

keyframe['frame_count'] must be a real int (bools explicitly rejected) describing the post-alignment frame count of the conditioned video.

Source

Thrown at python/sglang/multimodal_gen/runtime/pipelines_core/stages/model_specific_stages/minimax_h3/stages/denoising.py:106

            )
        return
    if keyframe is None:
        if task == "fl2va":
            raise ValueError("fl2va denoising requires encoded keyframe condition rows")
        return
    if not isinstance(keyframe, Mapping):
        raise ValueError("encoded keyframe condition rows must be a mapping")

    semantic_indices = tuple(keyframe.get("semantic_frame_indices") or ())
    if semantic_indices not in MINIMAX_H3_FL2VA_KEYFRAME_SIGNATURES:
        raise ValueError(
            "keyframe denoising requires semantic_frame_indices in "
            f"{MINIMAX_H3_FL2VA_KEYFRAME_SIGNATURES!r}, "
            f"got {semantic_indices!r}"
        )
    frame_count = keyframe.get("frame_count")
    if isinstance(frame_count, bool) or not isinstance(frame_count, int):
        raise ValueError("keyframe payload requires an integer frame_count")
    if frame_count <= 1:
        raise ValueError("keyframe payload frame_count must be greater than one")
    pixel_indices = keyframe.get("pixel_frame_indices")
    expected_pixel_indices = [
        frame_count - 1 if index == -1 else index for index in semantic_indices
    ]
    if pixel_indices != expected_pixel_indices:
        raise ValueError(
            "keyframe denoising requires pixel_frame_indices resolved from the "
            "semantic anchors, "
            f"got {pixel_indices!r} for frame_count={frame_count}"
        )

    entries = keyframe.get("keyframes")
    if (
        not isinstance(entries, list)
        or len(entries) != len(semantic_indices)
        or any(not isinstance(entry, Mapping) for entry in entries)

View on GitHub (pinned to 0132848349)

Solutions

1. Coerce with int(...) when building the payload (guarding against bool)
2. Convert numpy integers via int() before attaching
3. Ensure the field is always set by the producing stage

Example fix

# before
payload["frame_count"] = float(n_frames)
# after
payload["frame_count"] = int(n_frames)

fc = keyframe.get("frame_count")
assert not isinstance(fc, bool) and isinstance(fc, int)

def is_int_frame_count(v) -> bool:
    return not isinstance(v, bool) and isinstance(v, int)

When it happens

Trigger: frame_count missing (None), a float like 129.0, a string "129", or a boolean sneaking in from JSON/flag code.

Common situations: JSON deserialization producing strings/floats; numpy integer types not matching isinstance(int); a bool used as a 1/0 flag in the payload.

Related errors

• encoded keyframe condition rows must be a mapping
• keyframe payload frame_count must be greater than one
• keyframe denoising requires one encoded keyframe per semanti
• subblock_sparse_query_block_mask must be a tensor
• strict fl2va packed layout requires integer keyframe_frame_i

AI-assisted analysis of sgl-project/sglang@0132848349 (2026-08-28). Data as JSON: /api/errors/8b3f424699eacc1a. Report an issue: GitHub.