Prompt enhancer workflow not clear at all

[akabez]

Hey,

first of all from the little tests I have done I got to say the speed is amazing!
Still getting so so results but it's a start. Thank you for the models,

However the whole prompt enhancer is not clear at all.

On the code page it says see examples for more details on the prompt enhancer but there are no more details,
the image itself explains nothing really, to me at least.

If I start the workflow, it downloads something (the LLM models?) but where to? is it downloading an LLM model? how big is it?
After it downloads I need to specify in the node and point to it? but where was it downloaded to?
I have no LLM models folder setup for my comfyui and I couldn't find anything inside the LTX folder, is it all just automatic?
And then after it downloads everything I get an error that TIMM is not installed. No where do you mention that TIMM should be installed.

And lastly, the prompt enhancer text field input is extremely small and not very convenient to type into not to mention it cuts the text and shows only the start of the prompt (first 5 words max) not a major issue (yes copy paste I know) but not very convenient

It's a shame that sometimes developers do such an awesome work but than brush off the guides as an afterthought (I know you work hard but still) and forget that the MAJORITY of their users are not developers, you got to remember most people who USE gen AI are creatives not developers.

I don't want to sound ungrateful, I am not, just some constructive criticism
And of course thanks again for the model

Assaf

[Wolfven7]

are you using comfyui portable? I solved the timm issue on portable.

[rocky533]

Far as i know LLM models are placed in your models dir inside comfy, (one of the folders in there)
You can right click the prompt enhancer and select covert widget to input, then you can connect a larger text box to the prompt area.

[itistripleR]

Unlock the full potential of Lightricks (#LTXVideo) with my exclusive LTX + Gemini ComfyUI Workflow! As a huge thank you to the #LTXVideo team for their amazing open-source release, By integrating Gemini into ComfyUI, this workflow simplifies video generation, making it accessible for creators of all levels.

https://www.patreon.com/posts/ltx-gemini-123897759

[barclayslb]

are you using comfyui portable? I solved the timm issue on portable.

thanks, i solved it too with
pip install timm

[akabez]

Thank you all for replaying,
It's not really an issue per say, it just means you need to install TIMM.

The only thins I wanted to say is it's a shame developers do an amazing work on the product itself
but kind of neglect the instructions forgetting the majority of their users are not developers that use python and git all day every day if at all.

The TIMM thing was just another thing I mentioned, I was more talking about the fact the prompt enhancer workflow is not very clear when they say see more details below but there are no more details except the workflow image.

Anyway like I said it's not a rant or anything just some input.

Thanks again