"LoRA loading is experimental but it should work with just the built-in LoRA loader node(s).". But how to do it?

[a11y-a11y]

I can't find out how to load a GGUF lora.... see picture below:

There's no LoRA loader in GGUF nodes.
UNET loader can't be connected to anything that could load a LoRA.

Does anyone have a clue?

[kanttouchthis]

safetensors loras work just fine with the LoraLoaderModelOnly. you don't need a gguf lora. if you only have the gguf lora then you need to convert it back to safetensors

[a11y-a11y]

safetensors loras work just fine with the LoraLoaderModelOnly. you don't need a gguf lora. if you only have the gguf lora then you need to convert it back to safetensors

I have converted FLUX.2 LoRA to GGUF, and I need to know how to load it, so if it works, I can quantize it to Q8_0, thus avoiding the extra MEGATON RAM usage of normal safetensors LoRA. Is there a way to do it with this node?

[kanttouchthis]

safetensors loras work just fine with the LoraLoaderModelOnly. you don't need a gguf lora. if you only have the gguf lora then you need to convert it back to safetensors

I have converted FLUX.2 LoRA to GGUF, and I need to know how to load it, so if it works, I can quantize it to Q8_0, thus avoiding the extra MEGATON RAM usage of normal safetensors LoRA. Is there a way to do it with this node?

AFAIK gguf loras are not supported. comfyui supports fp8 lora which should be easy enough to convert. otherwise i would recommend fusing the lora into the gguf model (either fuse into safetensors model then convert to gguf or fuse into gguf model directly), since its a turbo lora this is actually a good idea since you will always be using it