Download Image&Mask

This is the combined version: one image_url field in, both an IMAGE and its MASK out. If you'd otherwise be chaining RiceRoundDownloadImageNode and RiceRoundDownloadMaskNode for the same source, this node collapses that into a single fetch - useful whenever the mask you need is the alpha channel of the same file rather than a separately hosted mask image.

Why this exists alongside the split versions

RiceRound's pack gives you both a combined node and two separate single-purpose ones (RiceRoundDownloadImageNode for image-only, RiceRoundDownloadMaskNode for mask-only). The split ones matter when your image and mask genuinely live at different URLs - a photo hosted one place, a separately generated mask hosted another. This node is for the more common case where they're the same file: fetch once, get both the pixels and whatever alpha data came with them, mirroring the relationship RiceRoundImageNode (RiceRound's local-file LoadImage equivalent) has between its own image and mask outputs - same idea, network fetch instead of a local file picker.

The input and outputs that matter

• image_url (STRING, no default) - the single URL this node fetches from. There's no separate mask URL field; the mask is derived from whatever alpha data the fetched image itself contains.
• Two outputs: image (IMAGE) and mask (MASK), both sourced from that one download.

Since RiceRound Publish is designed to read Input-category nodes like this one as fields on your generated page, this is a reasonable default choice whenever a published workflow needs to accept a user-supplied image with an alpha channel - an inpaint source photo, for instance - without you having to reason about whether the mask is a genuinely separate asset.

Installing it

Search "RiceRound" in ComfyUI Manager and install, or by hand:

cd ComfyUI/custom_nodes
git clone https://github.com/RiceRound/ComfyUI_RiceRound

then restart ComfyUI. No dependencies or model downloads are called out in the README for this pack - it's a packaging and network-fetch layer over a workflow you've already built, so install stays light.

Common issues & troubleshooting

The mask output is blank or all-white/all-black. That usually means the source image at the URL doesn't actually carry meaningful alpha data - a plain JPEG, for instance, has no alpha channel at all, so whatever this node derives as a mask from it won't be useful. If you need a real mask and your source format can't carry one, you're better off hosting the mask separately and using the split RiceRoundDownloadImageNode + RiceRoundDownloadMaskNode pair instead.

The download fails outright. Same generic URL-fetch issues as the single-purpose download nodes: an unreachable host, a link needing authentication, or a URL that resolves to an HTML page rather than a direct image file. None of this is specific to the combined node - it's the same network dependency either way.

Deciding whether it belongs in a workflow you're about to publish. It does - it's Input-category, which per the README is exactly the family of nodes meant to end up in the graph you feed into RiceRound Publish, unlike the separate Output-category debugging nodes.