|
--- |
|
license: creativeml-openrail-m |
|
tags: |
|
- coreml |
|
- stable-diffusion |
|
- text-to-image |
|
--- |
|
# Core ML Converted 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. |
|
- `split_einsum` version is compatible with all compute unit options including Neural Engine. |
|
- `original` version is only compatible with `CPU & GPU` option. |
|
- Custom resolution versions are tagged accordingly. |
|
- The `vae-ft-mse-840000-ema-pruned.ckpt` VAE is embedded into the model. |
|
- This model was converted with a `vae-encoder` for use with `image2image`. |
|
- This model is `fp16`. |
|
- 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).<br> |
|
|
|
# meinamix_meinaV10: |
|
Source(s): [CivitAI](https://civitai.com/models/7240/meinamix)<br> |
|
|
|
<img class="mantine-7aj0so" src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/b1522a8f-7e55-45b3-9be0-9af20de5bdfd/width=450/00011.jpeg" alt="00011.png" style="max-height: 100%; max-width: 100%;"> |
|
<img class="mantine-7aj0so" src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/c651d77a-e32e-415b-bcae-2750425be5dd/width=450/00015.jpeg" alt="00015.png" style="max-height: 100%; max-width: 100%;"> |
|
<img class="mantine-7aj0so" src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/8d2cc622-ad72-4724-aaa7-0c6316653494/width=450/00004.jpeg" alt="00004.png" style="max-height: 100%; max-width: 100%;"> |
|
<img class="mantine-7aj0so" src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/3e4ca1d9-c35e-4796-bdb2-943d7af15120/width=450/00005.jpeg" alt="00005.png" style="max-height: 100%; max-width: 100%;"> |
|
|
|
MeinaMix objective is to be able to do good art with little prompting. |
|
|
|
I created a discord server where you can post images that you generated, discuss prompt and/or ask for help. https://discord.gg/meinaverse |
|
|
|
You may also try my model using Sinkin.ai: https://sinkin.ai/m/vln8Nwr |
|
|
|
If you like one of my models and want to support their updates: |
|
I have a ko-fi and Patreon page where you can support me or buy me a coffee <3 , it will be very much appreciated: https://ko-fi.com/meina and https://www.patreon.com/MeinaMix |
|
|
|
MeinaMix and the other of Meinas will ALWAYS be FREE. |
|
|
|
The images in the samples aren't cherry picked, if i had spent 2h to cherry pick each of the images it would hide how the model really performs and i don't think thats fair. |
|
|
|
Recommendations of use: |
|
-------------------------------------------------------------------------------- |
|
Enable Quantization in K samplers. |
|
|
|
Hires.fix is needed for prompts where the character is far away in order to make decent images, it drastically improve the quality of face and eyes! |
|
--------------------------------------------- |
|
Recommended parameters: |
|
Sampler: Euler a: 40 to 60 steps. |
|
Sampler: DPM++ SDE Karras: 30 to 60 steps. |
|
Sampler: DPM++ 2M Karras: 20 to 60 steps. |
|
CFG Scale: 4 to 11. |
|
Resolutions: 512x768, 512x1024 for Portrait! |
|
Resolutions: 768x512, 1024x512, 1536x512 for Landscape! |
|
Hires.fix: R-ESRGAN 4x+Anime6b, with 15 steps at 0.4 denoising. |
|
Clip Skip: 2. |
|
Negatives: ' (worst quality, low quality:1.4), (zombie, sketch, interlocked fingers, comic), ' |
|
-------------------------------------------------------------------------------- |
|
|
|
In the merged models list: MeinaMix V1~6, MeinaPastel V3~5, MeinaHentai V2~3, Night Sky YOZORA Style Model, PastelMix and Facebomb, i do not have the exact recipe because i did multiple mixings using block weighted merges with multiple settings and kept the better version of each merge. |
