Doubt About image generation
Hello Everyone,
I was looking for a plugin that you can add a prompt and it can generate once click 20 or 30 in image instead just one image. Is there this tool or plugin in Huggin Face, please ?
Anyone Could you please help me ?
It's called comfyui. How do you not know about it. ya gotta at least search a bit before asking questions.
Thank yo so much for your help.
I tryed a lot after asking, I've found perchance, but it just have API that is not free.
I found other one that have one paid, no more for all life that is videoexpress, but there is no AI and I have to add text by text that make video with duration 7 seconds no more, but as I told I have input phrase by phrase one by one manually to generate the videos.
I really seach for, because of this I ask for the community
No offense, but it seems clear that English is not your primary language. Perhaps the language gap is the reason for your difficulties. Anyways; text to image is easy to do locally with webUIs like ComfyUI or others like SD Next or A1111, SDN is newer of the two, but only ComfyUI will get the more diverse features like Flux and audio/video generation. But, even with Comfy, audio/video is not necessarily an easy process. Download the comfyui manager to make it a bit easier, but even with that I'm personally having some difficulty implementing these things, and I also fear they may be attached to paid services, or reserved for people with 5-10'000$ PCs.. Still gonna keep trying until I find out though, but it's far from being plug and play. Do you use Windows or Linux?
Thank yo so much for your help.
I tryed a lot after asking, I've found perchance, but it just have API that is not free.
I found other one that have one paid, no more for all life that is videoexpress, but there is no AI and I have to add text by text that make video with duration 7 seconds no more, but as I told I have input phrase by phrase one by one manually to generate the videos.I really seach for, because of this I ask for the community
I stumbled upon your post again as I was going through my email and reread it. I'm not sure how I missed it the first time, but, did you say that comfyui is paid? Because it really isn't. A lot of these things have hookups for cloud services, but you can mostly ignore them. You just need your own models to use them, which can also be downloaded.
Most the models are free as well. The only thing is, you do need hardware good enough to run it all, and that's where the cost comes in when it comes to local generation. You technically can generate on as low as a 2gb Vram Nvidia GPU, but I would recommend at least 6gb VRAM(no less), if you want an experience that won't be frustrating.
If you only want to do inference (generation), and not actually train the models yourself, then 12-16 GB VRAM should be good for most of the best models and even since video generation. But if you want to train the models, the VRAM increases quite a bit, depending on what you want to train and at what quality.
PS: make sure Nvidia GPU. I hate to push brands, but that corporation has done some serious gatekeeping when it comes to AI, with their proprietary software/firmware. You technically can generate on other cards like AMD/Intel, but it goes much smoother on Nvidia cards.