Hugging Face's logo

coreml-community
/
coreml-meinamix_meinaV10

Text-to-Image
Core ML
stable-diffusion
Model card Files Community
coreml-meinamix_meinaV10
File size: 4,108 Bytes 
958d568
 
989321a
 
 
 
958d568
989321a
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
 
---
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.