Відео

It is possible to use my voice for this song creation?

The workflow is missing

I'm coming back to this workflow that I've mastered (finally!). It now allows me infinite possibilities and gives me ready prompts to prolong my experiences. For all this thank you again. You allow me through your work to live much better at the end of my life. Sincerely,

this actually works! thanks man. theres absolutely no way id figure this out--bless the node geniuses

I can't run on RTX 3070 Ti 8 GB+RAM 32GB error is torch.OutOfMemoryError: Allocation on device Got an OOM, unloading all loaded models. Prompt executed in 38.87 seconds

Very good tutorial thank you

Its truly insane what this thing does. I tried to upscale a aerial picture of my city and the details are very high but every house, car, trees, everything is changed but at the right place.

Music too high

thank you

Great video! Wished you showed an actual comparison of how ChatGPT would have answered since that was the bench mark you used in the beginning. For some of the benchmarks like the poem if you went a little more detail into the concepts of what a good poem looks like that would have been helpful.

how to install

I hope this message finds you well! I’m currently exploring AI and came across the term “AI flux., comfy UI” It piqued my interest, and I was wondering if you could share any insights, resources, or guidance on learning more about it. Wanted to learn from you, Can you please guide me..

Showing a video you created on 8gb vram is impressive, but that should be used as proof your workflow is fast, because i can have 8gb and create a video in any workflow if i tweak the settings, its just gonna take A LOT longer, what you should do is tell us the time it took to generate

I tried on a 3080ti gguf q6 takes up almost 12 GB I wanted to ask you if it is possible to increase the duration of the video

original approach. Have you tryed flux with an ultimateSDupscale with a controlnet blur. i don't remember the name of the controlnet model but when i'll come back home if you want i'll share

You are the best!)

awesome, will you make a similar workflow for LTX+ComfyUI ?

Hi, May I know where can I download ApplyFBCacheOnModel ?

I don't understand what is the work of upload image ?

Does this also work on saree

Are there local tools that would help improving the prompt to be detailed instead of using online chat bot?

bro how do you get the same face again and again. can you please help me out

When i try it with human photos the video is moving alot and im not getting that exact pose frame.

ты лучший чувак и спасибо за воркфлоу давай еще классных штук

hi, even after installing pulidflux i get an error Missing Node Types : PulidFluxModelLoader, PulidEvaClipLoader, PulidInsightFaceLoader, ApplyPulidFlux. Any idea why ? thanks 🙏

I'm going to give the ltx trick a try. I hope it works with partially blurred images also. Because using just img2img the blurry parts stays mostly blurry.

what is the background music? is it also generated? if so, how? can you make a video on it too

can u connect kokoro TTS model to this?

I could not get this workflow to produce anything worth snot and I do not understand why. I am using fp8 and q8 as I have a 4090.

Update the workflow. I downloaded the workflow and it's completely different to yours in the video.

SamplerCustomAdvanced The parameter is incorrect.

It's nice to see you

can you make your all workflow run on runpod how load missing models from civitai, huggingface. some time some models cant find, cant load

Great idea, great video breakdown

Can u post ur benchmark runs

Phi 4 is full of synthetic training data it doesnt have nsfw content u can use Sao 10k models

Can we use LMstudio instead of ollama It has better UI

do we need triton for wavespeed

Works on my RTX 2060🥹 but does not adhere to my prompts

Imo For accuracy Claude sonnet > gpt 4 ≈ minimax > Gemini 1.5

Great technique! One question: what if I need to put into my generated image some specific objects? For example, a very specific table lamp or a particular vase, or a poster on the wall with exactly specified text and graphics on it - can I use your workflow for that? Or can you recommend a more suitable approach for tasks of that kind? Thanks!

keep it up

Hi thanks for the video, did you include Florence to create prompts or can it also be used without Florence?

Impressive

good job, work fine

Thanks! Great explanation.

good to see you, keep up the good work

any way to run without gguf?

What version of guf should I download with a 12GB VRAM card?

Cannot execute because a node is missing the class_type property.: Node ID '#122' all is newest version