I built an agent system which will autonomously iterate & generate img of AI model wearing certain cloth and produce millions+ social posts
Free access to run any comfyUI workflow, hands-restoration model, upscaler & more on Replicate: https://replicate.fyi/ai-jason
🔗 Links
– Follow me on twitter: https://twitter.com/jasonzhou1993
– Join my AI email list: https://crafters.ai/
– My discord: https://discord.gg/eZXprSaCDE
– Get the source code: https://forms.gle/yYaE4WvzY7qHW3j16
– Install ComfyUI: https://github.com/comfyanonymous/ComfyUI
– Deploy your comfyUI workflow: https://replicate.fyi/jz-any-comfyui-workflow
– Hand-restoration model: https://replicate.fyi/jz-hand-restoration
– Ultimate-portrait-upscale: https://replicate.fyi/jz-ultimate-portrait-upscale
– Install insightface: https://www.youtube.com/watch?v=vCCVxGtCyho&ab_channel=DataLeveling
– IP Adapter (FACEID) Models: https://huggingface.co/h94/IP-Adapter-FaceID/tree/main
– IP Adapter Models: https://huggingface.co/h94/IP-Adapter/tree/main
– ControlNet (SD1.5) Models: https://huggingface.co/lllyasviel/ControlNet-v1-1/tree/main
– ControlNet (SDXL) Models: https://huggingface.co/lllyasviel/sd_control_collection/tree/main
– Checkpoint Models: https://civitai.com/
– More in-depth comfy UI tutorial by Data Leveling: https://www.youtube.com/watch?v=oYjEFHb–RA&t=3s&ab_channel=DataLeveling
⏱️ Timestamps
0:00 Intro
0:43 Use case
2:01 How does img generation works
5:37 ComfyUI Basic
07:21 Method 1: Integrate cloth into existing image
09:38 Method 2: Finetune model & IPAdapter
10:55 Install comfyUI manager, custom nodes & models
14:30 Step 1.1: Generate AI model face
19:51 Step 1.2: Embed cloth img
21:43 Step 1.3: Hand restoration & Upscale
24:00 Step 1.4: Deploy image generation pipeline
24:43 Overview of how to build self-improving agent system
26:12 Step 2.1 Setup agent tools to call AI models
29:30 Step 2.2 Setup multi agents system
34:02 Demo
👋🏻 About Me
My name is Jason Zhou, a product designer who shares interesting AI experiments & products. Email me if you need help building AI apps! [email protected]
#gpt5 #stablediffusion #openaidalle #autogen #gpt4turbo #gpt4 #ai #artificialintelligence #tutorial #stepbystep #openai #llm #chatgpt #largelanguagemodels #largelanguagemodel #bestaiagent #chatgpt #agentgpt #agent #autogen #autogpt #openai #aiimages
source
What kind of comfyUI image generation pipeline do you want me to build next?
@AIJasonZ can you add the code to description? Thank you for your knowledge sharing! 🙏
many great ideas, thx
This is really dope man
This channel is giving away free start up ideas and I absolutely love it!! ❤
As always awesome detailed video with perfection buddy :-)) in future Amazon and other similar websites will be using these projects to create model and put different clothes on them for showcasing instead of hiring models plus if any fork of this project with reduced complexity is developed in which people can put there faces too and see how they might actually look like easily while wearing particular pattern of clothes then many would like this feature I guess
Dude you are amazing!!! An absolute legend, a total beast, a god!
Hi, possible to have the final workflow comfyUI ? Thanks for the video
❤🔥
Sick! Now I want to play with comfyUI
I am pretty sure that Temu have been doing this for some time
I really want to get access to this as an API, i will pay
does this work on other accessories likes necklces, rings, and holding products by hand like perfumes, ointment ….
You’re an absolute boss, I hope you’re making $1 million a year
This is awesome AI Jason. Thanks a lot!
Jason, you are so amazing!
Interesting idea with using an agent for that! Just in time for me 🙂 I've trained a LoRA on my subject but the generating and upscaling take too much time. I only wonder how much does it cost with gpt4-v and running the model on Replicate? They both were quite expensive the last time I used them
Damn this comfy AI looks really awesome. Long live Open Source!!
This guy gives real life practical projects. Thanks as always!
Can comfyUI be uploaded to some server like vercel and use this image generation (on auto-pilot) or some kind of make trigger?
awesome workflow
@aijasonz Awesome work yet again AI Jason! Would love to know what other youtubers learn from as well as other sources you find to be helpful in growing your AI skills.
@aijasonz what computer would you recommend for doing all these ML tasks on? Anything affordable 🙂
I watched many videos which shows things in bits and pieces. But in this video you put all the bits and pieces together and present it with a real world example where the output is really real world standard. Absolutely amazing. This one video is really better than many ComfyUI playlist put together. I learnt a lot from this video. 🎉
comfyui >>> automatic1111
can you make a video on how to run many of these replicate models locally for free?
Wow, bravo. Nice job incorporating agents into the workflow
Mr. Jason, you are just amazing. I hope I can one day work with you on my project together. Keep going. 🙂 Many thanks.
One thing might has your attention, here is a breaking update on the ipadapter plus happened at 2024/03/23, the old ipadapter nodes used in this workflows won't work. FYA. Thanks for doing this demo, awesome.
Man's building entire venture backed companies in 30 min videos
This is absolutely incredible. I am extremely excited to be able to use this project. What a game changer. Thank you so much Rockstar!!
Good work ! Very Impressive
we shouldnt be filling the internet with even more garbage
Hey how to get the source code? Is there any github repo?
mind blown🙌 . i learnt comfy ui basics in one video
i came across best multi agent use case til date in the same video.
watched it twice at 2x and its the best time i spent on something this week. 🎉
keep them coming 🎉