You can do it using PuLID..I have tried it with ComfyUI but never tried it using code. Let me work on it and I will make a video on how we can do that..Thanks, @GenAILearniverse
I am going to make another video that how we can generate consistent characters. In the meanwhile, you can check PuLID on Google/YT to understand how it works. Thanks, GenAILearniverse
They stopped most stable diffusion and comfy from being launched. Only a hand full can generate picture using golab. You do need t4 for it to work properly
ERROR: pip's dependency resolver does not currently take into account all the packages that are installed. This behaviour is the source of the following dependency conflicts. torchaudio 2.5.1+cu121 requires torch==2.5.1, but you have torch 2.5.0 which is incompatible. torchvision 0.20.1+cu121 requires torch==2.5.1, but you have torch 2.5.0 which is incompatible. could you help me to fix that?
Hi, it seems like there is some conflict with the Cuda version you might have installed in some other project. I would suggest restarting the runtime in Colab and try running the first block again. If it doesn't work try logging in using some other Gmail Id. Please let me know if it helped. Thanks, GenAILearniverse
i am to error : ERROR: pip's dependency resolver does not currently take into account all the packages that are installed. This behaviour is the source of the following dependency conflicts. torchaudio 2.5.1+cu121 requires torch==2.5.1, but you have torch 2.5.0 which is incompatible. torchvision 0.20.1+cu121 requires torch==2.5.1, but you have torch 2.5.0 which is incompatible.
Yes Agree, thanks for pointing out. Google Colab will give you around 4 hours of runtime. Though you can try the 2nd option which is free as of now. Thanks, GenAILearniverse
Perfect, yes for the new users you get the discount...I have seen people buying the subscription from different Gmail Id rather than renewing the existing subscription to save some money. Thanks, @GenAILearniverse
I think the google colab is not completely free as when you check at the resources, it stated offered free is not gurantee and it has only given a limited runtime . Eg. 3 hrs & 50 mins. After you have used up the runtime, it will not be able to connect and generate image. Pls advice if this is not correct. Thank you.
Hi, You're correct that Google Colab has a limited free runtime (e.g., 3 hrs 50 mins), and usage is not guaranteed as it depends on resource availability. To extend runtime or access more resources, you can consider upgrading to Colab Pro or Colab Pro+. Thanks for pointing this out!...
I have been using Colab for a while now, for RUclips transcription. Yes, you can upgrade Colab if you want the runtime faster. I have transcribed more than 10 hours since I started and it still working
Hi, you can change Flux..here is another video I made using SDXL, please see if it helps. Currently it is built using local code but the same can be deployed on Google Colab. ruclips.net/video/vPtx_UfwhHA/видео.html Thanks, GenAILearniverse
Hi, as per my knowledge currently there are no completely free solutions available. There are text/image to video open source models but due to the huge size, it doesn't support the free Colab version. However the way GenAI is progressing, I am quite sure that we will have some quantized version available soon, that can be used on our local machine or on Colab to generate unlimited videos. As soon as something is available, I will make one video on that topic. Thanks, GenAILearniverse
Hi, The speed majorly depends on factors like Google Colab's server load and network bandwidth/internet speed. During peak times, Colab may allocate fewer resources, causing slower performance. Thanks, GenAILearniverse
Hi, let me make another video to explain how you can use Lora using Colab. In the meanwhile if you have a machine with good configuration, you can check this video...This is for using SDXL Lora : ruclips.net/video/r1nU6GPLe58/видео.html
You can do it using PuLID. Try checking on RUclips and you should be able to find some videos. I am also planning to make one for the same. Thanks, GenAILearniverse
Ohk, I do not agree much...it's completely free and allows you to run models that you cannot run on your local without having a very good configuration. Obviously you cannot use it for your production ready product but a good option if you try some POC or want to test something...Though if someone can afford it, they always have an option to buy a high end Machine with a GPU to run these models. Thanks, GenAILearniverse
I just need to learn how to use the prompts and settings. Im so bad at using AI that even after copying someone else's prompt and running it on the same platform, the result was nowhere near the original 😂 Not even in the same genre at that
Hi, don't worry, learning prompts is an iterative process! The beauty (and limitation 😊) of GenAI is that it doesn’t give the same results every time-it depends on how you see it. Take some ideas from the reference prompt, tweak the wording and parameters, and you should be able to get similar or even better results. Thanks, GenAILearniverse
Yes, I took the subscription since I had to generate multiple videos for this video. Though Kling will give you daily free credit. You can also use minimax that gives you an option to generate unlimited images for the first 3 days. Thanks, GenAILearniverse
Hi, you can use the 2nd option, I showed in the video or let's wait for another video wherein I will show another method using which you can generate unlimited images fast and absolutely free. Thanks, GenAILearniverse
cutting-edge supercar design, futuristic body panels with sharp angles, dramatic lighting highlighting powerful engine details, glossy carbon fiber accents
great tutorial, but how consistent character (face, hairs etc.) in different image settings can be created?
You can do it using PuLID..I have tried it with ComfyUI but never tried it using code. Let me work on it and I will make a video on how we can do that..Thanks, @GenAILearniverse
good video. how can you generate a consistent character using these image generators?
I am going to make another video that how we can generate consistent characters. In the meanwhile, you can check PuLID on Google/YT to understand how it works. Thanks, GenAILearniverse
Thanks bro ❤. Appreciate your work 💎🙏
They stopped most stable diffusion and comfy from being launched. Only a hand full can generate picture using golab. You do need t4 for it to work properly
Thanks so much , specially for the amazing prompts
Thank you
Thanks for the information, happy I found your channel 🎉
ERROR: pip's dependency resolver does not currently take into account all the packages that are installed. This behaviour is the source of the following dependency conflicts.
torchaudio 2.5.1+cu121 requires torch==2.5.1, but you have torch 2.5.0 which is incompatible.
torchvision 0.20.1+cu121 requires torch==2.5.1, but you have torch 2.5.0 which is incompatible.
could you help me to fix that?
Hi, it seems like there is some conflict with the Cuda version you might have installed in some other project. I would suggest restarting the runtime in Colab and try running the first block again. If it doesn't work try logging in using some other Gmail Id. Please let me know if it helped. Thanks, GenAILearniverse
i am to error : ERROR: pip's dependency resolver does not currently take into account all the packages that are installed. This behaviour is the source of the following dependency conflicts.
torchaudio 2.5.1+cu121 requires torch==2.5.1, but you have torch 2.5.0 which is incompatible.
torchvision 0.20.1+cu121 requires torch==2.5.1, but you have torch 2.5.0 which is incompatible.
but work create image😊
Misleading video. Google colab is not free unlimited. after using up the free time, you need to buy server time
Yes Agree, thanks for pointing out. Google Colab will give you around 4 hours of runtime. Though you can try the 2nd option which is free as of now. Thanks, GenAILearniverse
Excellent. Weldone. taught in a good simple way. Keep it up.
Glad you liked it.. Thanks, @GenAILearniverse
can i use this image for my youtube videos
Yes
@@GENAILearniverse thanks a lot
Thank you so much for sharing.. it’s awesome ❤
The credits In kling Ai, is it one time credit or does it gets added everyday?
It gets added everyday.
i have purchase on black Friday offer 10$ now it is showing 32$ , i got one time credit 3000.
Perfect, yes for the new users you get the discount...I have seen people buying the subscription from different Gmail Id rather than renewing the existing subscription to save some money. Thanks, @GenAILearniverse
I think the google colab is not completely free as when you check at the resources, it stated offered free is not gurantee and it has only given a limited runtime . Eg. 3 hrs & 50 mins. After you have used up the runtime, it will not be able to connect and generate image. Pls advice if this is not correct. Thank you.
Hi, You're correct that Google Colab has a limited free runtime (e.g., 3 hrs 50 mins), and usage is not guaranteed as it depends on resource availability. To extend runtime or access more resources, you can consider upgrading to Colab Pro or Colab Pro+. Thanks for pointing this out!...
I have been using Colab for a while now, for RUclips transcription. Yes, you can upgrade Colab if you want the runtime faster. I have transcribed more than 10 hours since I started and it still working
thanks bro very helpfull, can i change flux and models or vae? and how?
Hi, you can change Flux..here is another video I made using SDXL, please see if it helps. Currently it is built using local code but the same can be deployed on Google Colab. ruclips.net/video/vPtx_UfwhHA/видео.html
Thanks, GenAILearniverse
@GENAILearniverse thanks you👍👍
Hi bro, Please tell me Genai image to video generate for no hidden chargers and for unlimited generations.
Hi, as per my knowledge currently there are no completely free solutions available. There are text/image to video open source models but due to the huge size, it doesn't support the free Colab version. However the way GenAI is progressing, I am quite sure that we will have some quantized version available soon, that can be used on our local machine or on Colab to generate unlimited videos. As soon as something is available, I will make one video on that topic. Thanks, GenAILearniverse
@@GENAILearniverse
Thanks bro, i am eagerly wait for your video. Actually everyday i am checking your channel for if you upload kind of video.
Can we finetune it on our images on google colab bro?
Yes, you can use FluxGym. Search Lora training using FluxGym with Google Colab and you will find many videos on this topic..Thanks, @GenAILearniverse
It works very slowly today. What does it depend on?
Hi, The speed majorly depends on factors like Google Colab's server load and network bandwidth/internet speed. During peak times, Colab may allocate fewer resources, causing slower performance. Thanks, GenAILearniverse
how can we use lora in colab with flux ?
Hi, let me make another video to explain how you can use Lora using Colab. In the meanwhile if you have a machine with good configuration, you can check this video...This is for using SDXL Lora : ruclips.net/video/r1nU6GPLe58/видео.html
@@GENAILearniverse with sdxl i know how to use loras because of fooocus UI but with flux no . I don't have a good machine , i use colab PRO thou
How do we use it for consist character
You can do it using PuLID. Try checking on RUclips and you should be able to find some videos. I am also planning to make one for the same. Thanks, GenAILearniverse
is it legal ?
Yes, it is.
Thanks
this is an amazing site
Google cloab is shit. It's always out of computing power.
Ohk, I do not agree much...it's completely free and allows you to run models that you cannot run on your local without having a very good configuration. Obviously you cannot use it for your production ready product but a good option if you try some POC or want to test something...Though if someone can afford it, they always have an option to buy a high end Machine with a GPU to run these models. Thanks, GenAILearniverse
I just need to learn how to use the prompts and settings. Im so bad at using AI that even after copying someone else's prompt and running it on the same platform, the result was nowhere near the original 😂
Not even in the same genre at that
Hi, don't worry, learning prompts is an iterative process! The beauty (and limitation 😊) of GenAI is that it doesn’t give the same results every time-it depends on how you see it. Take some ideas from the reference prompt, tweak the wording and parameters, and you should be able to get similar or even better results. Thanks, GenAILearniverse
bro you have kling pro version
Yes, I took the subscription since I had to generate multiple videos for this video. Though Kling will give you daily free credit. You can also use minimax that gives you an option to generate unlimited images for the first 3 days. Thanks, GenAILearniverse
Thanks bro
Mistral ai add flux pro to its platform and its 100% free
Great, I didn't know that.. Thanks for sharing 👍
But I don't like its quality. For me krea is best.
740th...Thanks
I very like it
Thank you! 😊
make pulid flux for colab bro
Sure, this is on my to-do list..you will see it soon on my channel...Thanks
Takes too much time to generate image.
Hi, you can use the 2nd option, I showed in the video or let's wait for another video wherein I will show another method using which you can generate unlimited images fast and absolutely free. Thanks, GenAILearniverse