Add a LoRA to ComfyUI: verify the file and both model/CLIP branches

Purpose and scope

Start with one compatible LoRA and a controlled comparison. A populated selector does not establish that the sampling path uses the modified weights.

Source-supported context

The official example places LoRA Loader in both the model and text-encoder paths with separate strength controls. Adapter weights are not ordinary checkpoints. Source 1

Before you start

1. Establish a working baseline with a known base family and version.

2. Obtain the LoRA author's base-model, usage and artifact information.

3. Place it in this instance's loras category or the configured equivalent.

Steps and checks

The sequence below is an editorial procedure based on the cited context, not a diagnosis already confirmed for your environment.

Step 1. Verify source/completeness and select the real file after scanning. Resolve path problems if it is absent.

Step 2. Send the loader's MODEL to the actual sampler and CLIP to the corresponding text encoder, removing unintended bypass connections.

Step 3. Compare with/without one LoRA at a fixed seed and the author's suggested range, then vary model and CLIP strengths separately.

Step 4. Add further LoRAs one at a time, recording order, strengths and warnings. Investigate compatibility before reacting to key/shape failures.

Completion check

Graph connections and comparison results explain whether the LoRA participates, with the unmodified baseline available for recovery.

Limitations and cautions

Not all LoRAs alter the same layers or the text encoder. Weak visual effect alone is not installation failure, and exact character/style reproduction is not guaranteed.

Original sources

• LoRA workflow — original record checked 2026-09-21.

Localization prepared: 2026-09-22. The source-check dates above were inherited from the existing article; this translation does not claim they were all rechecked today. No GPU run, minimum-memory measurement or runtime guarantee is supplied.

Related troubleshooting and guides

• LoRA has no effect or reports size mismatch: check the base model and both output branches
• Value not in list: why a downloaded model does not appear in ComfyUI