Just my Comfy nodes. But feel free to use it. Prompt Database Node, LMStudio server access (prompt enhancer and more).
Installation in readme. (TL;DR) copy both (or just one) directories into Comfyui/Custom_Nodes.
These are nodes made with Codex, that do something i could not find, the way i wanted it, with simple use but some features i really wanted. Prime example the (intelligent) bypass option AND the option to edit the previous enhanced prompt, which is re-sent on bypass, IN the node directly. Without copy and pasting stuff around in text and clip nodes. And some Vram saving server options like loading/unloading the LMStudio model . Anyway:
Prompt Database:
Save multiple prompts (or whatever) with names, in three different categories. It's stored in the nodes folder, in json file (despite the "database" name) so it is available in any workflow in your comfy installation. Should be simple to understand.
LMStudio Prompt Enhancer:
Send any system prompt to a localhost server (Example LMStudio) and optionally a input prompt to enhance and/or a image (requires a vision model oc) and get the llm's answer back, exportable as string for any text input in Comfyui. Should be self explanatory as well. Doesn't need to be used for prompts, system prompt is completely free.
It auto removes any thinking block! But thinking still counts into the max tokens. So set them higher for thinking models. Especially on long, complicated system prompts.
Some caveats like that are covered in tooltips (Example to use the load and unload model switches, LMStudio (not me) requires a API token to be set.
But you don't need a token to just send it to whatever model the LMStudio server has currently loaded. Just disable the load/unload switches and it's in easy mode. the "require authentication" option needs to be disabled in LMStudios server settings of course (else you need a api token!).
Optionally put a "clear vram" node in front of it, so it kicks any comfy models out of vram before loading the lm studio model! Else LMStudio may try to laod that xxGB LLM model while LTX or whatever is still in it.
How to copy a model id to add to the node? :
remember you can leave it empty, leave load/unload unselected and so simply send it to whatever is loaded in the LMStudio Server!
Both in full display, Should be self explanatory, some fineties are covered by tooltips. This is as example Workflow under files.