Flux.1 IMG2IMG + Using LLMs for Prompt Enhancement in ComfyUI!

While we wait for the Flux.1 Dev controlnet models and ip adapters in ComfyUI, how about some composition control using image to image? Turns out that works just great, only often I find myself need longer prompts as they seem to generate better images. Or maybe if I could just ask for what I want, even if I’m not really sure? If only there was an AI to help with that?

Links, as shown in the video:

Want to support the channel? Get pre-made workflows and more!

== Beginners Guides for ComfyUI ==

[FlowFidelity]

2:41 That's how that works! Oh my goodness! I heard someone say the picture is the workflow, but didn't get it. Now I do :) Thank you!

[memoryhero]

The constant quick side commenting was magical in this vid - you kept it brief enough so that veterans won't feel bogged down by old redundant info but also that newbies will highly benefit from it. World class tutorial protocol.

[esuvari]

Canny for flux has just been released today

[Cadmeus]

A node that really helps to manage latent/image sizing btw is an underappreciated little extension from Ser-Hilary, called SDXL_sizing. Automatically spits out the right size for any base resolution (e.g. 512, 1024, 2048), at any given aspect ratio. I use wildcards from Impact Pack as an input to the aspect ratio setting, which works *really* well with Flux.

[Copperpot]

Excellent job on this workflow! Playing w/ it now after making a few of my own/using some common on civ/discord, but your incorporation of the LLM Party node + autosizing/etc is simply brilliant. Hope all is well!

[purposefully.verbose]

"nice beaver"
ok, thanks for that.

[akratlapidus]

Nerdy, you always deliver!!! 👌🏻👏🏻👏🏻👏🏻👏🏻

[Pygon]

Regarding the license, while the Output can be "used for commercial use", Flux has pretty clearly been trained on copyrighted material. While Flux (like other AI companies) might be able to get the fair use exceptions for the training that is currently being litigate across a number of cases, you would still likely be personally liable for every copyright infringement in the output. Those damages can be significant, so something just to keep in mind if your intention is to use the outputs commercially.

[wakegary]

Bill didn't seem to mind the multiples of 16! Stay nerdy!

[synthoelectro]

and for those who are stuck with 4GB VRAM, just use a large virtual memory and about 768 x 768, it takes up to 8 mins depending but hey, we did it before on 1.5. and SDXL, we can keep going, you can do this.

[jonmichaelgalindo]

Even if outputs were non-commercial, education and reporting are protected fair use. (Education and reporting are inherently commercial. Teachers and journos have to get paid.) Fair use is the doctrine all AI model training is built on. 😊

[icchansan]

Amazing, tutorials to make a custom lora soon?

[raducodreanu]

Hi, thanks for the awesome tutorial! How did you get the labels updated for the switches input (ex: "1_llm_enhanced", instead of "text1")?

[scarletsword]

I really like Flux, but I'm disappointed that I keep having to buy a new GPU to keep up with the demands of new AI art models. 😁

[dagkjetsa]

Hello. Great tutorial! :) Though... I am trying to follow your tutorial and get llama 3.1 up and running in comfyui, but it is not working. I am not sure what to put into the base_url field of the API Large Language Model Loader. Any help? Also, why are you using this node and not the "Local Large Language Model"?
Thanks

[erikjohnson]

Almost 50K subs. I'll do my part. (just subbed)

[equilibrium]

A face detailer workflow for flux would be really useful.

[michaelbayes]

The ability to load the flux models via "Load Checkpoint" thows up an error for me ...Could not detect model type of ... did you have this problem?

[joeb]

Do you have a copy of the final workflow that we can download?

[TobinatorXXL]

hello my LLM party problem: Error code: 500 - {'error': {'message': 'llama runner process no longer running: -1 ', 'type': 'api_error', 'param': None, 'code': None}}