|
--- |
|
license: creativeml-openrail-m |
|
tags: |
|
- coreml |
|
- stable-diffusion |
|
- text-to-image |
|
- not-for-all-audiences |
|
--- |
|
# Core ML Converted SDXL Model: |
|
|
|
- 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). |
|
- 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. |
|
- `original` version is only compatible with `CPU & GPU` option. |
|
- Resolution is the `SDXL` default of `1024x1024` UNLESS NOTED OTHERWISE. |
|
- This model requires macOS 14.0 or later to run properly. |
|
- This model is quantized to `8-bits`. |
|
- This model was converted with a `vae-encoder` for use with `image2image`. |
|
- Descriptions are posted as-is from original model source. |
|
- Not all features and/or results may be available in `CoreML` format. |
|
- This model does not have the [unet split into chunks](https://github.com/apple/ml-stable-diffusion#-converting-models-to-core-ml). |
|
- This model does not include a `safety checker` (for NSFW content). |
|
- This model can not be used with ControlNet. |
|
|
|
<br> |
|
|
|
# coreml-starlightXLAnimated-v3_SDXL_8bit: |
|
Source(s): [CivitAI](https://civitai.com/models/143043/starlight-xl-animated?modelVersionId=182077)<br> |
|
|
|
## Starlight XL 星光 Animated v3 |
|
|
|
### This is an SDXL base model converted and quantized to 8-bits.<br><br> |
|
|
|
 |
|
|
|
 |
|
|
|
 |
|
|
|
 |
|
|
|
## Starlight is a powerful 2.5D-like SDXL model designed for creating stunning concept art and illustrations. |
|
|
|
Key Features |
|
- Excels in anime, fantasy, and semi-realistic styles. |
|
- Designed for rich details and mesmerizing visuals. |
|
- Evolving based on community feedback. |
|
|
|
## How to Use |
|
|
|
Starlight shines over a large range but performs best with a LOWER CFG setting (3 to 5). Lower CFG provides finer, richer details, aligning with the model's intended goal. Higher CFG offers more color but may have increased artifacts. |
|
|
|
Example Prompt: 1girl, masterpiece, best quality, highly detailed, sharp focus, dynamic lighting, vivid colors, texture detail, particle effects, storytelling elements, narrative flair, 16k, UE5, HDR, subject-background isolation |
|
|
|
Example Prompt: 8k portrait enchanting depiction of an iridescent dress worn by a beautiful woman, intricately detailed, digital masterpiece, mesmerizing play of colors and light, ultimate elegance, exquisitely detailed shimmering fabric, hdr |
|
|
|
- Sampler: DPM++ |
|
- Recommended Steps: 25 to 65 (40) |
|
- CFG / Guidance: 3 to 5 (3.6) |
|
|
|
**I really hope you enjoy this SDXL model !** |
|
|
|
☕️ Consider supporting Starlight's development by [buying me a coffee !](https://ko-fi.com/chillpixel) |
