Ino Show Lora Config

The InoNodes pack loves to pass LoRA settings around as one JSON string - compact for wiring, but useless for eyeballing. Ino Show Lora Config takes that string and cracks it open into eleven named outputs, so you can actually read what a LoRA config contains, inspect it, or grab one field to feed somewhere else.

Think of it as the debug view for the pack's config format. You hand it a LoRA config JSON (from InoGetLoraConfig, InoCreateLoraConfig, or any of the download-config nodes), and it unpacks the fields onto the canvas. If you're new to this pack, it's also the fastest way to learn what a LoRA config contains - the outputs are the schema.

How it works

You pass a config string (multiline, so pasting JSON is fine) and it JSON-parses it, then returns each key as its own output. If the parse fails, it returns all-empty outputs and prints a log line rather than crashing your graph - a genuinely nice failure mode. The field it expects is a LoRA config with name, base_model, type, trigger words, a file reference, weights, and metadata.

Inputs and outputs

• enabled - pack-wide toggle; off returns empty outputs.
• config - the LoRA config JSON string.

The eleven outputs, in order:

• name - the LoRA's human-readable name (e.g. a character or style name).
• base_model - which base it was trained for (flux1dev, etc.). Worth checking: a Flux LoRA won't load on SDXL and this field is where you'd spot that.
• type - what it is (person, style, concept...).
• trigger_word / trigger_words - the activation token(s) to drop into your prompt.
• file - the model file reference.
• weight_type - the dtype/weight format.
• model_strength and clip_strength - the two strength floats (FLOAT outputs).
• description and tags - whatever metadata the config carries.

Where it wires in

Two honest uses. First, inspection: run a workflow that builds a LoRA config, tap this node in, and confirm the trigger word and strengths before generation. Second, extraction: if you only need the trigger word to build a prompt string, this node pulls it out without you having to hand-parse JSON. Most people reach for it the first way - it's a read node, not a control node.

Installing it

Part of ComfyUI-InoNodes. ComfyUI Manager → search "ComfyUI Ino Nodes" → install → restart. Manual:

cd ComfyUI/custom_nodes
git clone https://github.com/nobandegani/ComfyUI-InoNodes
cd comfyui_ino_nodes
pip install -r requirements.txt

Restart after. Requires inopyutils and a recent ComfyUI (V3 schema, v0.18.1+); no API keys.

Common issues

If the outputs come back empty with a "string_to_dict failed" log line, the config string isn't valid JSON - usually a stray quote or a pasted snippet that isn't the full object. And one subtle trap: the config keys are strength_model/strength_clip internally, but this node outputs them as model_strength/clip_strength. Don't go hunting for a mismatch - that's just the pack's naming. Finally, remember the trigger word only matters if it actually reaches your prompt; see it here, but then wire it into your positive prompt text.