Misc Parameters

Misc Parameters (the output node) is the smallest of the Searge output family - it unpacks essentially one thing from the parameters bundle: the LoRA strength. It's the read-back counterpart to the Misc Parameters input node, where you set the operation mode, prompt style, and LoRA strength.

It's part of SeargeSDXL, the SDXL base-plus-refiner workflow pack Searge released at SDXL 1.0's launch in 2023. It sits in Searge/_deprecated_/UI/Outputs - v3-era internal wiring replaced by the v4.x data stream - and it's not something you place on the canvas; it comes with the workflow.

How it works

The Searge workflow keeps all settings in one PARAMETERS bundle and uses output nodes to surface individual values. This one surfaces the LoRA strength so the LoRA-loading part of the graph knows how hard to apply the LoRA you selected in the Model Names node. It also passes the parameters bundle through so other output nodes can chain off it. That's the whole job - one value out, plus the pass-through.

The mode and prompt-style choices you also set in the Misc Parameters input node don't come out here as strings; they get turned into selector integers elsewhere (that's the Flow Control Parameters node's job). This output node specifically carries the LoRA strength.

The outputs that matter

• lora_strength (FLOAT) - how strongly the selected LoRA is applied.
• parameters (PARAMETERS) - passed through for chaining.

There's nothing to configure - the only input is the bundle, so the value is whatever you set upstream.

How to install it

ComfyUI Manager: search SeargeSDXL, install, restart. Manual: cd ComfyUI/custom_nodes && git clone https://github.com/SeargeDP/SeargeSDXL.git, then restart. Manual installs need python -m pip install opencv-python in ComfyUI's Python environment first, per the README, or the pack won't load. This node needs no models itself, and the LoRA strength only matters if you've actually selected and installed a LoRA.

Common issues

If your LoRA barely shows in the output, the likely reason is the default lora_strength of 0.2 - that's low for SDXL LoRAs, which often want 0.6–1.0. Raise it in the input-side Misc Parameters node; this output node only reports the value, it doesn't set it.

As with every node in this pack, the most common outright error is version drift between the workflow JSON and the custom nodes - the output nodes' exact shape changed across versions, so keep them matched. And the honest context: this is deprecated internal plumbing from a workflow built around SDXL's refiner, a pattern the community has largely moved past in favor of single fine-tuned checkpoints. It still runs fine; it's just a period piece.