Core ML Converted Model:

This model was converted to Core ML for use on Apple Silicon devices. Conversion instructions can be found here.
Provide the model to an app such as Mochi Diffusion Github - Discord 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.
vae tagged files have a vae embedded into the model.
Descriptions are posted as-is from original model source. Not all features and/or results may be available in CoreML format.
This model was converted with vae-encoder for i2i.
Models that are 32 bit will have "fp32" in the filename.

Note: Some models do not have the unet split into chunks.

Elldreth's Vivid Mix:

Source(s): CivitAI

This mixed model is a combination of my all-time favorites AND new-found favorites, including a very popular anime model mixed with Zeipher's F222, Dreamlike, and H&A's awesome 3DKX_1.0b! Lastly, to top it off I used howder's jomad model. Every single model in this mix are great on their own. This mix allows you to take advantage of combined concepts and produces some great images.

What's it good at?

Realistic portraits

Stylized characters

Landscapes

Fantasy

Sci-Fi

Anime

Horror

It's an all-around easy-to-prompt general purpose semi-realistic to realistic model that cranks out some really nice images. No trigger words required. All models were scanned prior to mixing and totally safe.

So what's the difference between Vivid and all my other models?

This model adds a lot more detail and realism to the images created with it and not just with portraits but landscapes as well. The other thing this model is better at is taking Textual Inversion embeddings. Lucid and Retro are both very resistant to TI Embeddings but Vivid is transformed very easily with a good embedding.

What are you waiting for? Go get some great results from simple prompts.

What's new in v2.0?

Wow wow wow.. two big model releases have kept me busy testing and prompting.

F222 was replaced by Hassan's newest model release

The new H&A 3DKX Update replaced the older version

wavymulder's portrait+ was added

Dreamlike was udpated in the mix as well

The end result is a lot more realistic and vivid outcome. I used the same prompt to generate the new preview images as were used in v1.0.