Data_sampleData

A well-built preset in this pack doesn't just bundle which model to load - it can carry sampling settings too: steps, cfg, which sampler, which scheduler. Data_sampleData is the node that reads those specific values back out of a context bus, so you can verify what a preset actually set, or reuse those exact numbers to drive a separate, vanilla KSampler elsewhere in your graph.

It's filed under the pack's Deprecated category. Apt_Preset's own README describes a dedicated "Basic sampler" node elsewhere in the pack that "packages ComfyUI's built-in sampler input ports into a single port" - that's the more likely current home for this functionality, even though it isn't part of this batch of nodes. If you're starting fresh, look there first.

How it works

Wire in a context, and it reads off four sampling-related fields that a preset or controller stack may have set: steps, cfg, sampler, and scheduler. There's nothing to configure on the node itself - no overrides, no options - it's purely a read tap into whatever the upstream bus already carries.

Worth noticing: the sampler and scheduler dropdowns mirror ComfyUI's full built-in lists, including newer entries like res_multistep and gradient_estimation - so this node stays current with whatever samplers your ComfyUI core version ships, rather than the pack maintaining its own separate list.

The inputs and outputs that matter

• context (optional, RUN_CONTEXT) - the bus to read from. Nothing is required, so an unwired node just produces empty/default outputs rather than erroring.
• Outputs: context (passed through), steps (INT), cfg (FLOAT), sampler (one of ComfyUI's full sampler list - euler, dpmpp_2m, res_multistep, uni_pc, and dozens more), and scheduler (simple, karras, sgm_uniform, beta, and others from ComfyUI's built-in scheduler set).

How to install it

Comes with the full pack. Via ComfyUI Manager: search "ComfyUI-Apt_Preset" → Install → restart. Manually:

cd ComfyUI/custom_nodes
git clone https://github.com/cardenluo/ComfyUI-Apt_Preset.git

Windows: run install.bat from the cloned folder through ComfyUI's embedded Python. Linux/Mac: activate your ComfyUI venv and install requirements from that folder by hand. Restart ComfyUI. No dependencies beyond the base pack.

Common issues & troubleshooting

Outputs are just showing defaults, not what you expected from your preset. Make sure context is actually wired to output coming from the loader/preset chain that set those sampling values - this node reads only what's in the bus it receives, it doesn't go looking for a preset by name.

Sampler or scheduler name doesn't match what you typed into a preset file. These are pulled from ComfyUI's live built-in lists at the time the graph runs, so a name has to match core's current spelling exactly - if you hand-edited a preset's .toml and used an old or misspelled sampler name, it won't resolve here.

It's deprecated - worth still using? Fine for reading values out of an existing setup or debugging what a preset configured. For new work, check the pack's "Basic sampler" node (per the README) first - it's built specifically to replace the old input-port sprawl this data-reading approach was working around.