Yes you're speaking with the 🤗 @huggingface 🧨 diffusers library personally
- Tweets 128
- Following 22
- Followers 2,877
- Likes 103
it's about to hit the top right corner
The Logo in Context Spaces demo + 🧨 diffusers implementation is here! 🖼️🏷️
In-Context LoRA + Image-to-Image + Inpainting → allow you to apply your logos to anything
huggingface.co/spaces/multim…
3
19
1,946
Google's MUSE: muse-model.github.io/ reproduced ✅
Why token-base image generation❓
- very under explored research
- needed for true multi-modal models
- better at style transfer
👉 huggingface.co/spaces/amused…
1
8
1,022
Also available in 🧨Diffusers
See: huggingface.co/stabilityai/s… for more info.
4 lines of code:
👇
Introducing SDXL Turbo: A real-time text-to-image generation model.
SDXL Turbo achieves state-of-the-art performance with a new distillation technology, enabling single-step image generation with unprecedented quality, reducing the required step count from 50 to just one.
The code, research paper, and weights for non-commercial use are now available on our website.
You can test SDXL Turbo on Stability AI’s image editing platform @Clipdropapp, with a beta demonstration of the real-time text-to-image generation capabilities.
Learn more: stability.ai/news/stability-…
3
8
43
6,129
Stable Diffusion XL now runs in 4 steps.
@SimianLuo, Yiqin Tan, @psuraj28 Daniel Gu distilled SDXL into a Latent Consistency LoRA.
Check it out with 🧨Diffusers v23.0:
👉 huggingface.co/latent-consis…
1
29
129
18,389
. @_segmind's SDXL distilled model is trending 📈
Great to see the first distillation of SDXL become so popular 🔥
👉 huggingface.co/segmind/SSD-1…
🚀 Introducing SSD-1B: Fastest text-to-image model
Segmind Stable Diffusion 1B is our latest open-source text-to-image model. It's 50% smaller and 60% faster than SDXL 1.0 and offers superior image quality. Read more here: blog.segmind.com/introducing…
1
5
22
4,395
. @Lykon4072's popular DreamShaper and Absolute- Realism models now available in 🧨Diffusers' format!
- DreamShaper XL huggingface.co/Lykon/dreamsh…
- DreamShaper v8
huggingface.co/Lykon/dreamsh…
- DreamShaper Inpainting
huggingface.co/Lykon/dreamsh…
- Absolute Realism
huggingface.co/Lykon/absolut…
👇
6
510
Midjourney is getting more open-source competition!
The amazing @deepfloydai has released IF, an open-sourced version of @google's Imagen.
Try it:
👉 huggingface.co/spaces/DeepFl…
Learn about it:
👉 huggingface.co/blog/if
3
13
51
13,679
Hey mom, that's me in the PyTorch blog! 🔦
just do `pip install --upgrade pytorch` and diffusers🧨 is faster out of the box! 💨
pytorch.org/blog/accelerated…
4
712
Open Source DALL-E 2 now on 🧨 diffusers!
In our latest 0.11.1 release we have the UnCLIPPipeline that works with #KakaoBrain's first large scale open unCLIP model🍭
Try it out on🤗Spaces: huggingface.co/spaces/kakaob…
And check the docs: huggingface.co/docs/diffuser…
2
53
280
37,420
A picture is worth 1000 prompts 🖼️
Paint-by-Example 🎨 is now an official pipeline of diffusers! 🧨
Try it out here: huggingface.co/spaces/Fantas…
10
30
3 out of the 5 Stable Diffusion v2 models are now integrated on my main branch 🧨
- The 512 model huggingface.co/stabilityai/s…
- The 768 model huggingface.co/stabilityai/s…
- The inpainting model huggingface.co/stabilityai/s…
New model, as easy to use me!
24
122
Versatile Diffusion by @humphrey_shi Labs is out on diffusers 🧨!
All-in-one architecture: text-to-image, image captioning, image variations, text variations in one model! Competitive to SOTA all these tasks 🗣
Model card: huggingface.co/shi-labs/vers…
Demo: huggingface.co/spaces/shi-la…
1
5
AltDiffusion by @BAAIBeijing is out on diffusers 🧨!
It is a new model trained from scratch in Chinese and English image captions - with the text-encoder of a new bilingual CLIP - and based on @stablediffusion
Model card: huggingface.co/BAAI/AltDiffu…
Demo: huggingface.co/spaces/BAAI/b…
2
2
14
So, @Microsoft has this cool open source text-to-image model with an innovate architecture (look mom no UNets 🚲) called VQ-Diffusion
Thanks to our contributor @williamLberman - it is now available on diffusers! Check out how it performs on our demo huggingface.co/spaces/patric…
4
24
157
Community pipelines are not included in our releases - however all of can be loaded with the custom_pipeline parameter
The custom pipelines available are here: github.com/huggingface/diffu…
1
3
Prompt Weighting 🏋️
Our contributor SkyTNT added prompt weighting to Stable Diffusion, allowing increasing or decreasing influence of words in the prompt: github.com/huggingface/diffu…
Left: Neon light illustration of a mecha robot
Right: Neon light (illustration:1.5) of a mecha robot
1
5
Imagic Editing for Stable Diffusion ✏️
Our contributor MarkRich also implemented the Imagic paper by
@bahjat_kawar et. al. that allows for editing images with text: you upload a single image, your prompt and et voilà: you get your image with the edits: github.com/huggingface/diffu…
1
1
3
Composable Diffusion ⛓️
Our contributor MarkRich implemented @nanliuuu (et. al) Composable Diffusion to Stable Diffusion using Diffusers: github.com/huggingface/diffu…
With that, you can use operators such as AND and NOT in your prompts
"Michael Cera AND Heisenberg from Breaking Bad"
1
2
16
My version 0.7.2 was released last week - maybe you heard about how I run #StableDiffusion up to 2x faster in all hardware? 🏎️💨
What you may not be aware of is that now it is even easier to swap schedulers! Here's how to do it with the new Euler ones contributed by @hlky__ 🤩
1
14
103
