I've been using Stable Diffusion on a MBP via invoke.ai. Are there recommendations for better parameterization of SD? I can never match the quality of the images I find on the internet even when using the same prompt and (seemingly) the same knobs (e.g., same Model like Euler A, etc). [edited for clarification]
Running Stable Diffusion XL 1.0 in 298MB of RAM
101–110 of 166 posts
Re: Running Stable Diffusion XL 1.0 in 298MB of RAM
#102I've been using Stable Diffusion on a MBP via invoke.ai. Are there recommendations for better parameterization of SD? I can never match the quality of the images I find on the internet even when using the same prompt and (seemingly) the same knobs (e.g., same Model like Euler A, etc). [edited for clarification]
Do they specify it's straight from the generator? The process videos I've seen start with "a girl standing in a green field" and then an hour plus of inpainting to fix hands, pose, etc.
Re: Running Stable Diffusion XL 1.0 in 298MB of RAM
#103Earlier quoted context omitted.
I love making fun of people that don’t understand percentages… wait, wat?
117. 472% of grade school students are unable to readily convert between fractions and percentages. 38.157% of informally provided statistics are made up on the spot under the assumption nobody will actually check.
You have a typo - it’s actually 83.157%
Re: Running Stable Diffusion XL 1.0 in 298MB of RAM
#104Earlier quoted context omitted.
Can you explain that context a little bit of Turing complete?
You can’t regulate the ownership of computing devices. It’s too generic. There are too many of them.
They could ban and phase out systems with unsecure bootloaders. That would go a long way. Many vendors have already locked down their boot process.
Re: Running Stable Diffusion XL 1.0 in 298MB of RAM
#105I've been using Stable Diffusion on a MBP via invoke.ai. Are there recommendations for better parameterization of SD? I can never match the quality of the images I find on the internet even when using the same prompt and (seemingly) the same knobs (e.g., same Model like Euler A, etc). [edited for clarification]
This is the best I've tried so far, but no mac support I don't think. Its a feature packed fork of Fooocus, which was developed by the orginal ControlNet dev. The quality you can get from small prompts is mind boggling:
https://github.com/MoonRide303/Fooocus-MRE
For base SD 1.5, I use Volta, because its fast: https://github.com/VoltaML/voltaML-fast-stable-diffusion/com...
Really good SD 1.5 image quality comes from gratuitous use of finetunes, LORAs, controlnet and other augmentations. So you can, say, trace a base image for structure, specify prompting in certain areas of the image and so on. InvokeAI is actually quite feature packed, and has lots of these augmentations hidden in the nodes UI, but Volta and other UIs also expose them more directly.
Re: Running Stable Diffusion XL 1.0 in 298MB of RAM
#106[flagged]
I see you're opting for AGPL on a codebase that is designed to be embedded as a library. Genuine question, what kind of user did you have in mind when you decided on this license?
Re: Running Stable Diffusion XL 1.0 in 298MB of RAM
#107I wonder if this could be accelerated with the Pi's onboard GPU somehow.
Is not not already using it? I thought ONNX had a GPU runtime the Pi could use.
Re: Running Stable Diffusion XL 1.0 in 298MB of RAM
#108Re: Running Stable Diffusion XL 1.0 in 298MB of RAM
#109I've been using Stable Diffusion on a MBP via invoke.ai. Are there recommendations for better parameterization of SD? I can never match the quality of the images I find on the internet even when using the same prompt and (seemingly) the same knobs (e.g., same Model like Euler A, etc). [edited for clarification]
I have the same experience with Invoke.ai or MochiDiffusion in the MBP M1. I can only match the quality of other images with Automatic1111 (https://github.com/AUTOMATIC1111/stable-diffusion-webui).
You’ll need more time and memory compared to Invoke or an Nvidia graphics card, but it’s not that bad: 1-2 s/it for an image in standard 512x768px quality, 14-20 s/it for an image in high 1024x1536px quality (Hires Fix).
Re: Running Stable Diffusion XL 1.0 in 298MB of RAM
#110I've been using Stable Diffusion on a MBP via invoke.ai. Are there recommendations for better parameterization of SD? I can never match the quality of the images I find on the internet even when using the same prompt and (seemingly) the same knobs (e.g., same Model like Euler A, etc). [edited for clarification]
Are you using custom weights? I'm assuming you are but there is a major difference between using the default RunwayML 1.5 weights and using a model finetuned for a specific purpose.
Generally the trade off is that any of the impressive finetuned models are far less generalizable then the default weights, but in practice this is not a big deal and the results can be a substantial improvement.