Upload README.md
Browse files
README.md
CHANGED
|
@@ -1,3 +1,55 @@
|
|
| 1 |
---
|
| 2 |
license: creativeml-openrail-m
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 3 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
---
|
| 2 |
license: creativeml-openrail-m
|
| 3 |
+
tags:
|
| 4 |
+
- coreml
|
| 5 |
+
- stable-diffusion
|
| 6 |
+
- text-to-image
|
| 7 |
+
- not-for-all-audiences
|
| 8 |
---
|
| 9 |
+
# Core ML Converted SDXL Model:
|
| 10 |
+
|
| 11 |
+
- This model was converted to [Core ML for use on Apple Silicon devices](https://github.com/apple/ml-stable-diffusion). Conversion instructions can be found [here](https://github.com/godly-devotion/MochiDiffusion/wiki/How-to-convert-ckpt-or-safetensors-files-to-Core-ML).
|
| 12 |
+
- Provide the model to an app such as **Mochi Diffusion** [Github](https://github.com/godly-devotion/MochiDiffusion) / [Discord](https://discord.gg/x2kartzxGv) to generate images.
|
| 13 |
+
- `original` version is only compatible with `CPU & GPU` option.
|
| 14 |
+
- Resolution is the `SDXL` default of `1024x1024`.
|
| 15 |
+
- This model was converted with a `vae-encoder` for use with `image2image`.
|
| 16 |
+
- This model is quantized to `8-bits`.
|
| 17 |
+
- Descriptions are posted as-is from original model source.
|
| 18 |
+
- Not all features and/or results may be available in `CoreML` format.
|
| 19 |
+
- This model does not have the [unet split into chunks](https://github.com/apple/ml-stable-diffusion#-converting-models-to-core-ml).
|
| 20 |
+
- This model does not include a `safety checker` (for NSFW content).
|
| 21 |
+
- This model can not be used with ControlNet.
|
| 22 |
+
|
| 23 |
+
<br>
|
| 24 |
+
|
| 25 |
+
# DreamShaper-XL1.0-Alpha2_SDXL_8-bit:
|
| 26 |
+
Source(s): [CivitAI](https://civitai.com/models/112902/dreamshaper-xl10)<br>
|
| 27 |
+
|
| 28 |
+
## This is an SDXL base model converted and quantized to 8-bits.
|
| 29 |
+
|
| 30 |
+
Finetuned over SDXL1.0.
|
| 31 |
+
|
| 32 |
+
Even if this is still an alpha version, I think it's already much better compared to the first alpha based on xl0.9.
|
| 33 |
+
|
| 34 |
+
Basically I do the first gen with DreamShaperXL, then I upscale to 2x and finally I do an img2img step with either DreamShaperXL itself, or a 1.5 model that I find suited, such as DreamShaper7 or AbsoluteReality.
|
| 35 |
+
|
| 36 |
+
What does it do better than SDXL1.0?
|
| 37 |
+
|
| 38 |
+
No need for refiner. Just do highres fix (upscale+i2i)
|
| 39 |
+
|
| 40 |
+
Better looking people
|
| 41 |
+
|
| 42 |
+
Less blurry edges
|
| 43 |
+
|
| 44 |
+
75% better dragons 🐉
|
| 45 |
+
|
| 46 |
+
Better NSFW<br><br>
|
| 47 |
+
|
| 48 |
+
|
| 49 |
+
|
| 50 |
+
|
| 51 |
+
|
| 52 |
+
|
| 53 |
+
|
| 54 |
+
|
| 55 |
+
|