ARTSY FLUX

Version 2

(release date - 10/28/2024)
The newest version of Artsy Flux improves on the advantages of version one and attempts to fix issues the original version had. Random messy noise has been greatly reduced in most contexts while maintaining the level of detail and creativity that the model is known for. Color grading has been enhanced substantially by training in properly color graded data, reducing outputs that appear to have a photo filter.

The capability to create anime in various styles has been expanded on. The model is now capable of 2D, 2.5D, and 3D anime.

Pretty happy with this version, so it may be some time until there is another. Enjoy!

(lower LoRA strengths if you're having compatibility issues)

Model Description

This Model is a sophisticated, carefully crafted merge of various self-trained LoRAs designed into introduce compositional and detail concepts into Flux.1 Dev in order to invoke the best aesthetics and qualities that Flux has to offer. It excels at realism with diverse environments and stylistic compositions, but is also capable of 2D imagery, adding a comparable amount of detail and artistic flair. Images produced have more cohesive color grading and lighting than most base Flux.1 Dev outputs.

This artfully curated merge allows you to get the best out of Flux WITHOUT the use of detailer LoRAs.

Achieve compositions and levels of detail that can even rival Flux Pro at the fast speeds of any version that works best for your local configuration:

VERSIONS AVAILABLE:

FP16 (.safetensors)
FP8 (.safetensors)
Q8_0 (.gguf)
Q4_K_S (.gguf)

VERSIONS THAT WILL NEVER BE AVAILABLE (Do not ask):
NF4
AIO (clip_l, t5xxl and vae included)

These models use the same clip_l, t5xxl, and ae.safetensors as Base Flux.1-Dev.

(Images attached to each version have been generated with their respective model sizes/quants for accuracy and honesty. Though you'll be surprised at the level of detail achievable with this Q4_K_S.)

ENJOY!