fn_registry/python/functions/ml/diffusers_load_lora.md

name, kind, lang, domain, version, purity, signature, description, tags, uses_functions, uses_types, returns, returns_optional, error_type, imports, params, output, tested, tests, test_file_path, file_path

name	kind	lang	domain	version	purity	signature	description	tags	uses_functions	uses_types	returns	returns_optional	error_type	imports	params	output	tested	tests	test_file_path	file_path
diffusers_load_lora	function	py	ml	1.0.0	impure	def diffusers_load_lora(pipe: Any, lora: LoraRef) -> Any	Carga un adaptador LoRA en un pipeline diffusers via pipe.load_lora_weights. Si lora.weight != 1.0, aplica set_adapters para escalar la contribucion del LoRA.	diffusers ml lora image-generation fine-tuning		lora_ref_py_ml		false	error_go_core	diffusers	name desc pipe Pipeline diffusers que soporte load_lora_weights (SD1.5, SDXL, etc.). name desc lora Referencia al adaptador LoRA. lora.path al .safetensors o directorio. lora.weight escala la fusion (1.0 = completo, 0.5 = mitad).	El mismo pipe con el LoRA cargado y peso aplicado. Modificacion in-place, retorna pipe para composicion.	false			python/functions/ml/diffusers_load_lora.py

Ejemplo

from diffusers_load_lora import diffusers_load_lora
from lora_ref import LoraRef

lora = LoraRef(path="/path/to/my_lora.safetensors", weight=0.8)
pipe = diffusers_load_lora(pipe, lora)

Notas

Usa pipe.load_lora_weights(path) para cargar. Si lora.weight != 1.0:

• Intenta pipe.set_adapters(["default"], adapter_weights=[weight]) (diffusers >= 0.20).
• Fallback a pipe.fuse_lora(lora_scale=weight) para versiones antiguas.

El campo lora.scale (override de alpha) no se aplica aqui — diffusers no expone un parametro directo equivalente en la API publica actual. Se puede setear via pipe.load_lora_weights(path, weight_name=...) si el archivo tiene nombre especifico.

Import lazy de diffusers — ImportError descriptivo si no instalado.