Flux GGUF models are our favorite!

If you are just getting into FLUX, check out this simple workflow

Download the workflow:

Flux GGUF Models:

Clip models:

Flux VAE:

Flux Realism LORA:

[xiao_slayer]

omg a tutorial that actually tells us how and where to do thing instead of telling us to go download and figure it out. great work

[Huang-ujrt]

For a beginner also say that your videos are really very friendly, thank you very much. Because of my professional needs and the high learning threshold of flux, I've been using mimicpc to run flux before, it can load the workflow directly, I just want to download the flux model, and it handles the details wonderfully, but after watching your video, I'm using mimicpc to run flux again finally have a different experience, it's like I'm starting to get started! I feel like I'm starting to get the hang of it.

[AI-Rogue]

Got it working. Quality is amazing. Thank You!

[shuntera]

Generated in 100 seconds on my Nvidia 2080 Ti - works really well!!

[abrarrafat]

Best explanation and Guide GGUF right now. Can U describe the difference between various quantized models Such as Q8, Q4 ? Also, can it generate image in any resolution or does it have to specifically 1024x1024 ?

[ovworkshop]

Im so glad you didn't say "Is taking American by Storm like other videos are doing". 🤭

[lofigamervibes]

You have got to be from Florida... Not asking to be sus, just I know your accent so well lol... I'm def gonna try out the GGUF!

[jlyn]

Wow these ai voice filters are starting to sound pleasant to listen to

[amrokas]

Not recomend using GGUF with control net. It is way slower than safetensors. If I'm correct it's ~30% slower. Basic image generation with GGUF is same as with safetensors.

[magimyster]

Wow! Let's hope the models get optimized a bit more... Unfortunately, I work with just a CPU, and it takes me an hour to generate an image... 😑

[Avalon]

So with my RTX 3070 8GB took 7 minutes and that was without the upscaler, yep they really need to figure out optimization for these models, they are wonderful and great but they are so impractical for us little guys to use.

[moebiusSurfing]

Are this GGUF models useful also having an RTX 3090 with 24GB VRAM?
Maybe at the end of tweaking you can use the bigger models?

[Avalon]

I'm always hesitant lately to use these workflows with my RTX 3070 8GB machine, even more so I can't imagine tacking on a upscaler, I'll give it a try and see what happens and I also would like to know how you got the times on each node?

[contrarian]

How do you display the execution times on the nodes?

[bobtahar]

Why there is Lora Key Not Loaded in the cmd prompt?

I got similar issue... I don't think lora is working GGUF yet.

[gpr]

Thanks for the video.
How can I have the seconds be displayed above the nodes?

[sorijin]

I noticed there's a triple clip loader included in the gguf stuff, any good combo/use to using 3 over 2?

[DanielSchweinert]

Nice video but if you have 4090 and you need 40sec per generation something is not right. It should be more like 15sec per generation. I changed the t5xxl_fp8_e4m3fn to t5xxl_fp16 and it is faster.

[sulemanwaheed]

I am getting a missing custom node red box error for UNETLoaderGGUF and Any ideas where I can find these?

[Aitoolstation]

How much laptop requirement is necessary to run comfy UI ?