Use this LoRA with flux1.d-fp8
Set the strength to 1.0
steps 20
trigger words “nene” is not necessary
If you buzz me, I will use it to train more models.
flux_lora_nene_rank64_bf16.safetensors