AI images
- Thread starter pwe
- Start date
Kharzette
Watcher of Overs
- 4,964
- 3,628
Since my download is so slow I thought I would make some baldur's gate 3 taverny AI art! Prepare thyself for mangled fingers and really bad chainmail. All elves depicted are over the age of 400.
This one was a funny accident. I put "glasses of wine":
Edit: One more because it is cute:
This one was a funny accident. I put "glasses of wine":
Edit: One more because it is cute:
Last edited:
- 4
GuardianX
Perpetually Pessimistic
- 6,765
- 17,071
chthonic-anemos
bitchute.com/video/EvyOjOORbg5l/
- 8,606
- 27,267
- 30,474
- 22,323
Blazin
this is probably your best path in, assuming you have access to a decent GPU.Easy Diffusion 3.0 is now available to @everyone, and is easier, faster and more powerful than ever. It adds full support for SDXL, ControlNet, multiple LoRAs, Embeddings, seamless tiling, and lots more.
It also includes a bunch of memory and performance optimizations, to allow you to make larger images, faster, and with lower GPU memory usage. And the UI continues to remain easy-to-use and simple, even after adding the new features. And runs on Windows, Linux and Mac.
Easy Diffusion 1.0 was born at 3am on Aug 24, 2022, with 200 lines of code. One year later, the project has grown wildly beyond my expectations, with more than half a million installs, 18,000+ members in the Discord community, and a massive codebase (42,000+ lines of code in v3.0).
I'm very thankful to the amazing contributors and volunteers in the community who've helped keep Easy Diffusion alive and growing, despite the tiny team and lack of resources.
To use version 3.0, please restart Easy Diffusion, and it'll update automatically. Otherwise, please download from Release v3.0 - SDXL, ControlNet, LoRA, Embeddings and a lot more! · easydiffusion/easydiffusion to use our 1-click installers.
You can download the SDXL model from https://huggingface.co/stabilityai/...e-1.0/resolve/main/sd_xl_base_1.0.safetensors and save it in the models/stable-diffusion folder. Don't use an older VAE with this.
For fun: here's the first commit (one year ago): Initial commit · easydiffusion/easydiffusion@4fbd347 (edited)
- 3
Blazin
Creative Title
- 6,413
- 33,667
I'll look at it, been playing with Adobe firefly last few days which has been decent from my uninformed perspective.
Easy Diffusion 3.0 is now available to @everyone, and is easier, faster and more powerful than ever. It adds full support for SDXL, ControlNet, multiple LoRAs, Embeddings, seamless tiling, and lots more.
It also includes a bunch of memory and performance optimizations, to allow you to make larger images, faster, and with lower GPU memory usage. And the UI continues to remain easy-to-use and simple, even after adding the new features. And runs on Windows, Linux and Mac.
Easy Diffusion 1.0 was born at 3am on Aug 24, 2022, with 200 lines of code. One year later, the project has grown wildly beyond my expectations, with more than half a million installs, 18,000+ members in the Discord community, and a massive codebase (42,000+ lines of code in v3.0).
I'm very thankful to the amazing contributors and volunteers in the community who've helped keep Easy Diffusion alive and growing, despite the tiny team and lack of resources.
To use version 3.0, please restart Easy Diffusion, and it'll update automatically. Otherwise, please download from Release v3.0 - SDXL, ControlNet, LoRA, Embeddings and a lot more! · easydiffusion/easydiffusion to use our 1-click installers.
You can download the SDXL model from https://huggingface.co/stabilityai/...e-1.0/resolve/main/sd_xl_base_1.0.safetensors and save it in the models/stable-diffusion folder. Don't use an older VAE with this.
For fun: here's the first commit (one year ago): Initial commit · easydiffusion/easydiffusion@4fbd347 (edited)
- 11,144
- 42,182
So using this, feeding it my linkedin pic, and giving it the prompt "This man as Santa Claus"... I get this...
Easy Diffusion 3.0 is now available to @everyone, and is easier, faster and more powerful than ever. It adds full support for SDXL, ControlNet, multiple LoRAs, Embeddings, seamless tiling, and lots more.
It also includes a bunch of memory and performance optimizations, to allow you to make larger images, faster, and with lower GPU memory usage. And the UI continues to remain easy-to-use and simple, even after adding the new features. And runs on Windows, Linux and Mac.
Easy Diffusion 1.0 was born at 3am on Aug 24, 2022, with 200 lines of code. One year later, the project has grown wildly beyond my expectations, with more than half a million installs, 18,000+ members in the Discord community, and a massive codebase (42,000+ lines of code in v3.0).
I'm very thankful to the amazing contributors and volunteers in the community who've helped keep Easy Diffusion alive and growing, despite the tiny team and lack of resources.
To use version 3.0, please restart Easy Diffusion, and it'll update automatically. Otherwise, please download from Release v3.0 - SDXL, ControlNet, LoRA, Embeddings and a lot more! · easydiffusion/easydiffusion to use our 1-click installers.
You can download the SDXL model from https://huggingface.co/stabilityai/...e-1.0/resolve/main/sd_xl_base_1.0.safetensors and save it in the models/stable-diffusion folder. Don't use an older VAE with this.
For fun: here's the first commit (one year ago): Initial commit · easydiffusion/easydiffusion@4fbd347 (edited)
Apparently I am the three headed Santa which guards the gate to the North Pole.
- 4
- 2
- 1
- 32,088
- 125,795
- 30,474
- 22,323
- 11,133
- 41,590
- 30,474
- 22,323
Very little with Midjourney. Getting Stable Diffusion running locally is a bit of a chore, but infinitely more flexible once you do.
- 1
- 11,133
- 41,590
I just downloaded Stable Diffusion and my anti-virus software is blocking it for some reason. I'll give Midjourney a shot.