invoke-ai / invoke-ai/InvokeAI

[enhancement]: LoRa - Populate triggers words automatically, color coding, grouping.

Open
#8,784 2 comments 2 reactions 1 assignee Claimed by @lstein View on GitHub
enhancement good first issue
Dominant language
Python
Stars
28.2k
Forks
3k
Avg merge
6d 5h
Merged PRs (30d)
19

Description

### Is there an existing issue for this?

- [x] I have searched the existing issues

### Contact Details

_No response_

### What should this feature add?

I've downloaded quite a few LoRA models for poses and characters, before i figured out you can add the prompts directly to the LoRA itself to easily add them in the prompt box.

So i thought that it would be much nicer if the models i downloaded from Civitai could automatically populate itself with the exact name it had on the website and also populate itself with the trigger words listed on the website. To avoid manually doing it all because the loras are named after the file name, which is often not name it was given on the website.

Another thing that i would like is being able to group the loras. Currently i'm "grouping" them by just prefixing them with something like "Pinup - " or "Character - "
So having an expandable box in the LoRa picker to easily go right to whatever group i made would be awesome.

Lastly, I think color coding the LoRa and their corresponding triggers words in the prompt menu would make it a but easier to see which triggers belong to what, or to see if i forgot a trigger of a lora i have active.

Below's a mockup of the color coding i had in mind.
Each LoRA gets a little color band on the header that one could edit, and the prompts would receive the same background olor.
![Image](https://github.com/user-attachments/assets/ae771cac-426f-4c68-a3cc-21a0ac95a10a)

### Alternatives

_No response_

### Additional Content

_No response_

Contributor guide

No contributing guide indexed for this repository

Assessment

This issue has not been assessed yet.

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.