Comfy-Org/ComfyUI · error · ValueError

geometry has no sky output; run with Mono/Metric models.

Error message

geometry has no sky output; run with Mono/Metric models.

What it means

DA3 geometry decode node raises this when output='sky_mask' is requested but the geometry dict has no 'sky' entry. Sky maps are only emitted by Mono/Metric DA3 models, so the requested output simply does not exist for the loaded model.

Source

Thrown at comfy_extras/nodes_depth_anything_3.py:437

            apply_sky_clip = output["apply_sky_clip"]
            if apply_sky_clip and "sky" not in da3_geometry:
                raise ValueError(
                    "apply_sky_clip=True requires a sky tensor in the da3_geometry input, but none is present. "
                    "Run with Mono/Metric models or set apply_sky_clip=False."
                )
            depth = da3_geometry["depth"]
            sky = da3_geometry.get("sky")
            if apply_sky_clip and sky is not None:
                depth = torch.stack([
                    da3_preprocess.apply_sky_aware_clip(depth[i], sky[i])
                    for i in range(depth.shape[0])
                ], dim=0)
            grey = cls._depth_to_image(depth, sky, normalization)  # (B,H,W,3) greyscale
            result = _turbo(grey[..., 0]) if output_val == "depth_colored" else grey

        elif output_val == "sky_mask":
            if "sky" not in da3_geometry:
                raise ValueError("geometry has no sky output; run with Mono/Metric models.")
            sky = da3_geometry["sky"]
            if output["colored"]:
                result = _turbo(sky)
            else:
                result = sky.unsqueeze(-1).expand(*sky.shape, 3).contiguous()

        elif output_val == "confidence":
            if "confidence" not in da3_geometry:
                raise ValueError("da3_geometry has no confidence output; run with Small/Base models.")
            conf = _normalize_confidence(da3_geometry["confidence"])
            if output["colored"]:
                result = _turbo(conf)
            else:
                result = conf.unsqueeze(-1).expand(*conf.shape, 3).contiguous()

        else:
            raise ValueError(f"Unknown output mode: {output_val}")

View on GitHub (pinned to 1c6d8d45b3)

Solutions

1. Switch the output to 'depth'/'depth_colored'/'confidence' as supported by your model.
2. Or load a Mono/Metric DA3 model, which produces sky geometry.

if output_val == 'sky_mask' and 'sky' not in da3_geometry:
    raise ValueError('this model has no sky output; use depth or confidence instead')

def geometry_has_sky(geo: dict) -> bool:
    return 'sky' in geo

When it happens

Trigger: Selecting 'sky_mask' as the output while the upstream DA3 node ran a Small/Base model (or any mode that doesn't populate the 'sky' key).

Common situations: Output dropdown left on 'sky_mask' after a checkpoint swap; following a Mono-model sky-segmentation workflow with a Small model.

Related errors

• apply_sky_clip=True requires a sky tensor in the da3_geometr
• da3_geometry has no confidence output; run with Small/Base m
• multi-view mode requires Small or Base model. The loaded mod
• pose_method='cam_dec' requires a camera decoder, but the loa
• pose_method='ray_pose' requires a DualDPT head, but the load

AI-assisted analysis of Comfy-Org/ComfyUI@1c6d8d45b3 (2026-08-14). Data as JSON: /api/errors/4de66fe1a9e8c1fb. Report an issue: GitHub.