Flux Dev To Schnell 4 Step LoRA - FP32 API Documentation
by ModelsLabThis version is float32 if your card doesn't support BF16 it's better to have this convert to FP16 than BF16. Less precision loss that way.
flux-dev-to-schnell-4-step-lora-fp32Deploy Dedicated GPU server to run AI models
Deploy ModelThis version is float32 if your card doesn't support BF16 it's better to have this convert to FP16 than BF16. Less precision loss that way.
flux-dev-to-schnell-4-step-lora-fp32