Debargha Bhattacharjee

Пікірлер

@unitedstrafes4522 7 күн бұрын

Can't get any controlnet stuff to work in forge+flux (have it running fine in every other model) for some reason getting it to work in forge plus flux eludes me. Any advice? 🤷‍♂

@muhammadtalha9401 10 күн бұрын

Is there any software which can do image to video OR text to video

@Debargha_Bhattacharjee 9 күн бұрын

kzbin.info/www/bejne/iJSxlmScnNGWsMk You can start from here. There is also runaway which is another great tool for this.

@Maastertips 14 күн бұрын

Can you suggest a website for making stock images?

@Debargha_Bhattacharjee 9 күн бұрын

Midjourney, dall-e, leonardo ai there all are amazing now.

@Thee.Mighty 16 күн бұрын

does not work, no models

@Debargha_Bhattacharjee 15 күн бұрын

Can you specify the issue you are facing?

@petertremblay3725 18 күн бұрын

I get way better speed in forge, i use this recent model ''fluxFusionV24StepsGGUFNF4_V2GGUFQ5KS'' that is blazing fast and result are on par with FP8 or FP16 at 4 steps.I also use turbo_Alpha lora and my gpu is an RTX 3060 12 GB and i can generate an image at 800X600 in 22 sec in forge and in comfyui the same setup take 148 sec. So after reading a bit on the issue it seem like the forge dev are very good at optimisation.

@Debargha_Bhattacharjee 15 күн бұрын

That's a huge difference. Thanks for pointing it out. I am definitely gonna try this tonight.

@petertremblay3725 14 күн бұрын

@@Debargha_Bhattacharjee You wont regret it enjoy!

@Ai_mayyit 19 күн бұрын

hey debargha unable to install this node layer diffuse it is giveng error and error again and again do you have any fix for this?

@Debargha_Bhattacharjee 15 күн бұрын

Can you share which error it is giving?

@Ai_mayyit 15 күн бұрын

@@Debargha_Bhattacharjee it solve but not working because i wana remove car background and the background is remove but not from the car windows do you have any solution for removing car window background

@Flexeditz444 Ай бұрын

😮😮😮 fake

@Debargha_Bhattacharjee Ай бұрын

What do you mean by fake?

@VerseVilla-saj Ай бұрын

A new gig idea💡- reminder A date reminder

@wolf8088 Ай бұрын

thanks bhai simple bhabe bojhanor jonno , besir vaag video creator ghuriye firiye katha bole...

@Debargha_Bhattacharjee Ай бұрын

Really appreciate this comment.

@Anime_edits_ms Ай бұрын

Very very thankyou

@Debargha_Bhattacharjee Ай бұрын

You're most welcome.

@kmnNmk-je8jk Ай бұрын

how to run nf4 flux in python code in colab t4?

@Debargha_Bhattacharjee Ай бұрын

Sadly I don't have much idea about running it from collab.

@tenamsb686 Ай бұрын

Dude, if your pc has a 12 gb graphics card then it's not a potato pc..

@Debargha_Bhattacharjee Ай бұрын

I thought so too buddy. But the way things are going, I don't think this would be relevant for long given how much time it needs to generate a single image with Flux Dev. By the way Flux NF4 runs with even as low as 6 GB of V-ram, so that's what the title is about.

@Epulsenow 2 ай бұрын

Bhai Jaag Gaya, Coma main tha august se

@Debargha_Bhattacharjee 2 ай бұрын

😂😂😂 So true

@MaxPayne-o8f 2 ай бұрын

so k utha hai kya bhai 3 mahine se chalra hai forge me flux

@Debargha_Bhattacharjee 2 ай бұрын

Kind of 😂 Was stuck in some problem, so was inactive for 3 months. Now when I am back, so much has happened 😮🤷🏽

@a64738 2 ай бұрын

Updated stable defusion aaand now the LORA tab is totally gone... Tried disable all extension and turn of LORA and on again. Did not help.... But after update it had changed default settings to all extensions disabled which overrides that you turn them on or of. So I disabled that and then all extensions was back.

@jonhhex-k9s 2 ай бұрын

bhai step by step btao

@Debargha_Bhattacharjee 2 ай бұрын

More detailed videos coming through. Keep supporting ❤️

@EasyGAMEee 2 ай бұрын

Спасибо брат из Индии

@Flatwoodsdad 2 ай бұрын

You stated the problem well and your answer was shown just as well. Thank You.

@Debargha_Bhattacharjee 2 ай бұрын

You're welcome 😃

@umairkadri5383 2 ай бұрын

Can you suggest some free ai like midjourney to do this and also if this field is so full of people then is it even worth uploading?

@Debargha_Bhattacharjee 2 ай бұрын

Right now, I do not think this is a good idea going into this field. It is already very saturated. If you are planning on building a career using ai image generation tools, I would suggest going some other route. If you just want to have fun and do something on side, you might give this a try.

@Maltebyte2 2 ай бұрын

Thank you so much my friend!

@Debargha_Bhattacharjee 2 ай бұрын

I am happy that it was helpful for you.

@moheedkhan6656 2 ай бұрын

Paid version hai

2 ай бұрын

So easy, thank you very much !

@ChrissyAiven 2 ай бұрын

Hi Thanks, was lookign for that. But your first SD solution also works for me with SDXL 1.0 models without the second modification?

@nixonmanuel6459 3 ай бұрын

Thank you!

@Debargha_Bhattacharjee 3 ай бұрын

I'm happy that it was helpful.

@zetra021 3 ай бұрын

excellent, thanks a lot!

@Debargha_Bhattacharjee 3 ай бұрын

😃😃

@JinShinPlayground 3 ай бұрын

Unfortunately, I can't get it to work.

@Debargha_Bhattacharjee 3 ай бұрын

Can you explain the problem? I can try to help if I can.

@sandhyachalla7131 4 ай бұрын

Is the any dimension specification, to upload in adobe stock?

@Debargha_Bhattacharjee 3 ай бұрын

As per my knowledge, there is really no dimension specification in adobe stock. But you should always keep in mind the use cases of the images you would be uploading. If you are uploading a 2:1 ratio image which is actually suitable to be used in a square format or vice versa, it won't sell well. Also, the higher the image dimension the better.

@vipulthakkar1111 4 ай бұрын

Are you achive this permonth 2000 per month

@DigitalAI_Francky 4 ай бұрын

Hello, in ComfyUI there are 35 folders in my models, where are they in your paths ? I only see 6 paths but I have much more folders in my ComfyUI Models folders like unet for example, etc...

@Debargha_Bhattacharjee 4 ай бұрын

The folder shown in this video is the folder of Stable Forge, not the comfyui. So, the number of folders might seem different.

@paulohcassis 4 ай бұрын

Thank you for the video. I've installed the Stable Diffusion WebUI but couldn't make it work. Apparently, it's missing a Stable Diffusion checkpoint. What can I do to fix that? Your help is much appreciated.

@Debargha_Bhattacharjee 4 ай бұрын

You can download the checkpoints from huggingface or civitai and paste them inside the 'webui'>'models'>'checkpoints' folder. That would do the job.

@paulohcassis 4 ай бұрын

@@Debargha_Bhattacharjeethank you! It's working now. However the Layerdiffusion plugin is not working. I don't see it on the menu as you do. I know it is installed, but don't have the option to select it. Any idea?

@RiazEdits 4 ай бұрын

Allow to commercialy sale?

@Debargha_Bhattacharjee 4 ай бұрын

Stable diffusion has different policies for their different models regarding commercial usage. It's best to check their website for permissions regarding the specific model or model group you want to use.

@justthecoolestmusicjohne6105 4 ай бұрын

Can't understand a word you are saying.

@robertaopd2182 5 ай бұрын

can be this setup for flux ? somehow

@neelicious Ай бұрын

I second this!

@MusicFlowDaily 5 ай бұрын

Thank you! You saved me! :D

@Debargha_Bhattacharjee 5 ай бұрын

That's great to hear.

@RiiahTV 5 ай бұрын

thankyou i searched for years

@Debargha_Bhattacharjee 5 ай бұрын

I'm happy that it could be useful

@muertetotal7 5 ай бұрын

Thank you very much, you are a crack

@iAdden 5 ай бұрын

Does this work on Mac?

@Debargha_Bhattacharjee 5 ай бұрын

Absolutely. Comfyui works in both windows and mac. So do the workflows.

@iAdden 5 ай бұрын

@@Debargha_Bhattacharjee thanks

@TheAmit4sun 5 ай бұрын

Awesome . thanks man , that was easy.

@Debargha_Bhattacharjee 5 ай бұрын

Glad that it was helpful and easy to understand.

@Thecatanomia 6 ай бұрын

Aapni to dekhchi, reply i den na

@Debargha_Bhattacharjee 6 ай бұрын

Sorry for the late reply. It's been a very hectic schedule for last few days.

@Thecatanomia 6 ай бұрын

Goolge collab er sohit video banao, bro

@Debargha_Bhattacharjee 6 ай бұрын

Sure, I shall look into that.

@Thecatanomia 6 ай бұрын

Please bhai Ki kore animated animals video banabo ? Please dekhao, aami goolge collab use kori

@Debargha_Bhattacharjee 6 ай бұрын

Definitely. I am currently planning a video on a similar topic. Stay tuned.

@Thecatanomia 6 ай бұрын

@@Debargha_Bhattacharjee Thanks Bhai, need only Google Collab ones, please make a playlist of that, if you can 🙌

@DebarghaBhattacharjeeDB 6 ай бұрын

@@Thecatanomia I haven't used Google collab for sd yet. Once I wanted but then Google announced that they are banning all ai applications from collabs, so never bothered. I shall try to learn that part.

@Thecatanomia 6 ай бұрын

@@Debargha_Bhattacharjee Ok bro, NP ! Will try to learn from your videos then. Thank you 😊

@Thecatanomia 6 ай бұрын

Tumi to dekhchi vai, khub valo video banao

@Debargha_Bhattacharjee 6 ай бұрын

Thanks a lot.

@murphylanga 6 ай бұрын

cool video, thank you very much, I have to test this and integrate it into my automatic workflow😍

@sudabadri7051 6 ай бұрын

I know this is good if the legend murphy is excited

@Debargha_Bhattacharjee 6 ай бұрын

Glad to know that it was helpful. :D

@murphylanga 6 ай бұрын

@@sudabadri7051 Thank you, that flatters me. But I'm not a legend, I'm still alive. ;-) But you're right, it's good. I've already incorporated some of the nodes and will probably put the workflow and a video online at the weekend.

@litonray8183 6 ай бұрын

how to fix this File "E:\Ai Softwear\ComfyUI_windows_portable\ComfyUI\main.py", line 80, in <module> import execution File "E:\Ai Softwear\ComfyUI_windows_portable\ComfyUI\execution.py", line 11, in <module> import nodes File "E:\Ai Softwear\ComfyUI_windows_portable\ComfyUI odes.py", line 21, in <module> import comfy.diffusers_load File "E:\Ai Softwear\ComfyUI_windows_portable\ComfyUI\comfy\diffusers_load.py", line 3, in <module> import comfy.sd File "E:\Ai Softwear\ComfyUI_windows_portable\ComfyUI\comfy\sd.py", line 5, in <module> from comfy import model_management File "E:\Ai Softwear\ComfyUI_windows_portable\ComfyUI\comfy\model_management.py", line 119, in <module> total_vram = get_total_memory(get_torch_device()) / (1024 * 1024) ^^^^^^^^^^^^^^^^^^ File "E:\Ai Softwear\ComfyUI_windows_portable\ComfyUI\comfy\model_management.py", line 88, in get_torch_device return torch.device(torch.cuda.current_device()) ^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "E:\Ai Softwear\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\cuda\__init__.py", line 778, in current_device _lazy_init() File "E:\Ai Softwear\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\cuda\__init__.py", line 284, in _lazy_init raise AssertionError("Torch not compiled with CUDA enabled") AssertionError: Torch not compiled with CUDA enabled

@captainpike3490 6 ай бұрын

I have found and downloaded a similar workflow that includes Fooocus V2 prompt expansion. However, I couldn't get it to work. Could you please improve this workflow by incorporating Fooocus V2 prompt expansion and multiple LoRA options? Thank you.

@Debargha_Bhattacharjee 6 ай бұрын

As per my knowledge, the v2 effects the prompt creation in a completely different way than the base fooocus version. It is more like a tool for dynamically linking different prompts and styles to get the desired results. Can you share the workflow you got so that I can try to dissect it and try to combine them if possible? You can mail me at [email protected]

@captainpike3490 6 ай бұрын

@@Debargha_Bhattacharjee Already e-mailed to you. Thank you.

@zraieee 6 ай бұрын

well done

@Debargha_Bhattacharjee 6 ай бұрын

Thank you

@RealNikhilChoudhary 6 ай бұрын

This is the video i was looking for long long time........thankyou

@Debargha_Bhattacharjee 6 ай бұрын

Your comment prompted me to create this video. Thank you.

@The-Qwertyuiop123 6 ай бұрын

Please someone help. I have followed all of these steps, installed all dependencies, tried that model and many more, tried different images, and different tile sizes and it DOESN'T UPSCALE! The output looks exactly as ugly as the input, it resizes it but doesn't upscale. I have also tried uninstalling and reinstalling, but nothing works.

@alexolascoaga 6 ай бұрын

Same here

@Debargha_Bhattacharjee 6 ай бұрын

This upscaling method changes the pixel density but does not change how the image looks. If you want an upscaling method that changes how the image looks and creates newer details like magnific AI, I shall be making a video about that.

@The-Qwertyuiop123 6 ай бұрын

@@Debargha_Bhattacharjee Thank you very much! I'll be sure to watch it!

@The-Qwertyuiop123 6 ай бұрын

@alexolascoaga I've found that it actually does make a difference, smoothing out a lot of the noise, but you won't notice until you expand the image in the node with the dots in the lower right. It just isn't making the image details less pixely. Adding a denoise node also helps smoothen it out.

@The-Qwertyuiop123 6 ай бұрын

@alexolascoaga i tested many apps and the superimage app also upscaled to 4k very well!

@golpo_shikari 6 ай бұрын

amazing

@Debargha_Bhattacharjee 6 ай бұрын

Thank you

@golpo_shikari 6 ай бұрын

@@Debargha_Bhattacharjee dada khub valo kaaj korchen , tai apnar viewer ba view barbe na , eta mone kore niee egiee jan , ami sadharon manus , amar khomota nei , but apnar songe achi

@Debargha_Bhattacharjee 6 ай бұрын

Thanks a lot for your kind words. It means a lot and gives motivation for new videos.

@RealNikhilChoudhary 6 ай бұрын

please make a video on how to use fooocus models in comfiui

@Debargha_Bhattacharjee 6 ай бұрын

kzbin.info/www/bejne/mWG2nIV5g7WHi9Usi=q_Hgzu5RQV9qU3oI Thanks for commenting. Here you go.

Ең жақсы KZbin

Пікірлер