glid-3-xl
meadowrun
glid-3-xl | meadowrun | |
---|---|---|
7 | 2 | |
255 | 93 | |
- | - | |
0.0 | 9.1 | |
almost 2 years ago | 10 months ago | |
Python | Python | |
MIT License | MIT License |
Stars - the number of stars that a project has on GitHub. Growth - month over month growth in stars.
Activity is a relative number indicating how actively a project is being developed. Recent commits have higher weight than older ones.
For example, an activity of 9.0 indicates that a project is amongst the top 10% of the most actively developed projects that we are tracking.
glid-3-xl
-
Run Your Own DALL·E Mini (Craiyon) Server on EC2
If you’re anything like us, though, you’ll feel compelled to poke around the code and run the model yourself. We’ll do that in this article using Meadowrun, an open-source library that makes it easy to run Python code in the cloud. For ML models in particular, we just added a feature for requesting GPU machines in a recent release. We’ll also feed the images generated by DALL·E Mini into additional image processing models (GLID-3-xl and SwinIR) to improve the quality of our generated images. Along the way we’ll deal with the speedbumps that come up when running open-source ML models on EC2.
-
[D] Has anyone trained the latent diffusion models by OpenAI(CompVis)? Need some help
Personally, I've found GLID3 and GLID3-XL to be nice and straightforward. Worked right out of the box.
- [D] Any relatively new text2image models with fine tuning?
-
Trying to remember the name of an upscaler. I thought it was Glide XL or something.
GLID-3-XL?
-
Waiting for Dall-e 2's release feels like waiting for a birthday gift.
I've been using the Glid-3 XL model. Here are some examples of what I've been able to do with it:
-
Cyberpunk Anime Girl progressive enhancement
This was made with Glid-3 XL
-
"Woah there, Dragonman!" (16 output images with CompVis latent diffusion)
The above notebooks use GitHub repo GLID-3-XL from Jack000. Regarding CLIP guidance, Jack000 states, "better adherence to prompt, much slower" (compared to classifier-free guidance).
meadowrun
-
Run Your Own DALL·E Mini (Craiyon) Server on EC2
If you’re anything like us, though, you’ll feel compelled to poke around the code and run the model yourself. We’ll do that in this article using Meadowrun, an open-source library that makes it easy to run Python code in the cloud. For ML models in particular, we just added a feature for requesting GPU machines in a recent release. We’ll also feed the images generated by DALL·E Mini into additional image processing models (GLID-3-xl and SwinIR) to improve the quality of our generated images. Along the way we’ll deal with the speedbumps that come up when running open-source ML models on EC2.
-
Why Starting Python on a Fresh EC2 Instance Takes Over a Minute
So it is more reasonable to cache the download locally for up to 4 hours. That saves us 5–10 seconds on every run.
What are some alternatives?
latent-diffusion - High-Resolution Image Synthesis with Latent Diffusion Models
dalle-playground - A playground to generate images from any text prompt using Stable Diffusion (past: using DALL-E Mini)
dalle-flow - 🌊 A Human-in-the-Loop workflow for creating HD images from text
warehouse - The Python Package Index
glide-text2im - GLIDE: a diffusion-based text-conditional image synthesis model
glid-3-xl - 1.4B latent diffusion model fine tuning
dalle-playground - A playground to generate images from any text prompt using DALL-E Mini and based on OpenAI's DALL-E https://openai.com/blog/dall-e/
jax - Composable transformations of Python+NumPy programs: differentiate, vectorize, JIT to GPU/TPU, and more
meadowrun-dallemini-demo - A demo of using Meadowrun to run DALL·E Mini, GLID3-XL, and SwinIR in an image generation pipeline
distribution - Placeholder repository to allow filing of general bugs/issues/etc against the Clear Linux OS for Intel Architecture linux distribution