Comfy-Org/ComfyUI · error · ValueError

SeedVR2TemporalChunk: chunking_mode must be 'auto' or 'manua

Error message

SeedVR2TemporalChunk: chunking_mode must be 'auto' or 'manual'; got {mode!r}.

What it means

SeedVR2TemporalChunk reads the chunking mode out of the chunking_mode primitive dict and requires exactly 'auto' or 'manual'. This mirrors the UI combo; any other string — stale value, casing, or typo from a hand-built prompt — is rejected.

Source

Thrown at comfy_extras/nodes_seedvr.py:473

    def execute(cls, latent, temporal_overlap, chunking_mode) -> io.NodeOutput:
        samples = latent["samples"]
        if samples.ndim != 5:
            raise ValueError(
                f"SeedVR2TemporalChunk: expected a 5-D video latent (B, C, T, H, W); "
                f"got shape {tuple(samples.shape)}."
            )
        if samples.shape[1] != SEEDVR2_LATENT_CHANNELS:
            raise ValueError(
                f"SeedVR2TemporalChunk: expected {SEEDVR2_LATENT_CHANNELS} latent channels; "
                f"got shape {tuple(samples.shape)}."
            )
        if temporal_overlap < 0:
            raise ValueError(
                f"SeedVR2TemporalChunk: temporal_overlap must be >= 0; got {temporal_overlap}."
            )
        mode = chunking_mode["chunking_mode"]
        if mode not in ("auto", "manual"):
            raise ValueError(
                f"SeedVR2TemporalChunk: chunking_mode must be 'auto' or 'manual'; "
                f"got {mode!r}."
            )
        t_latent = samples.shape[2]
        t_pixel = 4 * (t_latent - 1) + 1

        if mode == "auto":
            free_gb = comfy.model_management.get_free_memory(
                comfy.model_management.get_torch_device()) / (1024 ** 3)
            mpx_per_frame = (samples.shape[0] * samples.shape[3] * samples.shape[4]) * (BYTEDANCE_VAE_SPATIAL_DOWNSAMPLE ** 2) / 1e6
            budget_gb = free_gb - SEEDVR2_CHUNK_RESERVED_GIB - SEEDVR2_CHUNK_SIGMA_K * SEEDVR2_CHUNK_SIGMA_GIB
            chunk_latent_max = max(1, int(budget_gb / (SEEDVR2_CHUNK_GIB_PER_MPX_FRAME * mpx_per_frame)))
            frames_per_chunk = min(4 * (chunk_latent_max - 1) + 1, t_pixel)
            logging.info(
                "SeedVR2TemporalChunk auto: free=%.2fGiB, %.2fMpx -> frames_per_chunk=%d (t_pixel=%d).",
                free_gb, mpx_per_frame, frames_per_chunk, t_pixel,
            )
        else:

View on GitHub (pinned to 1c6d8d45b3)

Solutions

1. Set chunking_mode to 'auto' or 'manual' exactly (lowercase).
2. Re-save the workflow in the current UI so the combo emits a valid value.
3. When building prompts programmatically, validate against {'auto','manual'} before queueing.

Example fix

# before
mode = {'chunking_mode': 'Automatic'}

# after
mode = {'chunking_mode': 'auto'}

mode = chunking_mode['chunking_mode']
if mode not in ('auto', 'manual'):
    raise ValueError(f"chunking_mode must be 'auto' or 'manual', got {mode!r}")

def is_valid_chunk_mode(v) -> bool:
    return v in ('auto', 'manual')

When it happens

Trigger: Supplying chunking_mode = {'chunking_mode': 'Auto'}, 'automatic', or a removed mode name via the API or an edited workflow JSON.

Common situations: Workflows saved by older/newer ComfyUI versions where the combo values changed; scripts injecting arbitrary strings; case mismatches.

Related errors

• SeedVR2PostProcessing: unknown color_correction_method {colo
• SeedVR2TemporalChunk: expected a 5-D video latent (B, C, T,
• SeedVR2TemporalChunk: expected {SEEDVR2_LATENT_CHANNELS} lat
• SeedVR2TemporalChunk: temporal_overlap must be >= 0; got {te
• SeedVR2TemporalChunk: frames_per_chunk must be a 4n+1 pixel-

AI-assisted analysis of Comfy-Org/ComfyUI@1c6d8d45b3 (2026-08-14). Data as JSON: /api/errors/5adb52ac57181132. Report an issue: GitHub.