A version of FLUX.2 [klein] 4B-base that supports fast fine-tuned lora inference
Want to make some of these yourself?
