nodes/TeaCrab-ComfyUI-TeaNodes
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Update README.md

Browse Source
This commit is contained in:
Brent Liu
2026-04-13 21:30:57 -04:00
committed by TeaCrab
parent 877f9c64c4
commit 1fb6b78294
1 changed files with 11 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files
README.md
+11
View File
@@ -35,6 +35,7 @@ Adds a few new nodes:
* Hopefully the color input field works for most color string formats.
### Random Lora & Model
<<<<<<< HEAD
* Uses regex pattern to filter down files within `models/checkpoints` and `models/loras` and then randomly choose one to load.
* `every` parameter allows the chosen model to ran certain number of times.
* `pause` parameter allows the chosen model to run indefinitely.
@@ -43,3 +44,13 @@ Adds a few new nodes:
* If a model has been used before in the same session, the node won't need to access the file from SSD/HDD again, it's already in the RAM.
* `RESULT` outputs the working state of the node, use `Preview As Text` or `Show Text` node to understand time until next randomization, which model is being used currently, previously and how many times, or which ones hasn't been used yet from the found files.
* If the regex pattern went wrong and there are no models found, a random model will be selected from all available under the respective `checkpoints` or `loras` folder.
=======
* Uses regex pattern to filter down files within `models/checkpoints`, `models/loras` directory and then randomly choose one to load.
* `every=#` parameter allows the chosen model to run only `#` number of times until randomly selecting another.
* `pause=True` allows the current running model to run indefinitely.
* `skip=True` or any changes in `pattern` will trigger a random selection ASAP, `skip` doesn't stop loop counting, but changes in `pattern` will.
* Models are pooled, which can lead to maxing out the RAM usage, Python seem to be able to handle this on its own without issue.
* If a model has been used before in the same session, the node won't need to access the file from SSD/HDD again, it's already in the RAM.
* `RESULT` outputs the working state of the node, use `Preview As Text` or `Show Text` node to understand time until next randomization, which model is being used currently, previously and how many times, or which ones hasn't been used yet from the found files.
* If the regex pattern went wrong or if there are no models found, a random model will be selected from all available under the respective `checkpoints` or `loras` folder.
>>>>>>> 0f38763 (Update README.md)