These are the midjourney styles that are pre-loaded in Whatchamacallit
Using original textual inversion bins that are compatible with most webuis/notebooks that support text inversion loading. They can be easily converted to diffusers-style and in Whatchamacallit there is code to do that already if you need reference.
- midj-strong:
good at that weird surreal melty almost golden sort of style, looks like clip guided diffusion in my opinion
- midj-portrait:
a bit more subtle but still very cinematic and changes the image significantly but less so than midj-strong
- midj-anthro:
was finetuned on some anthropomorphic animals (not traditional furry style, but just animals standing like humans). good on other subjects though.
Inference Providers
NEW
This model is not currently available via any of the supported Inference Providers.
The model cannot be deployed to the HF Inference API:
The model has no library tag.