fn_registry/python/functions/ml/diffusers_generate.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_generate	function	py	ml	1.0.0	impure	def diffusers_generate(pipe: Any, cfg: GenerationConfig) -> ImageGenResult	Ejecuta inferencia con un pipeline diffusers usando GenerationConfig. Mide duracion y pico de VRAM. Retorna ImageGenResult con imagen PIL, meta y metricas.	diffusers ml image-generation inference vram metrics pendiente-usar	genconfig_to_diffusers_kwargs_py_ml	generation_config_py_ml image_gen_result_py_ml	image_gen_result_py_ml	false	error_go_core	torch diffusers	name desc pipe Pipeline diffusers cargado y listo para inferencia (resultado de diffusers_load_pipeline, opcionalmente con scheduler y LoRA configurados). name desc cfg Parametros de generacion. cfg.seed >= 0 para semilla fija; -1 usa time-based. cfg.sampler se incluye en meta pero no se aplica aqui (usar diffusers_set_scheduler antes).	ImageGenResult con image=PIL.Image.Image, meta={backend, model, sampler, actual_steps, seed, width, height, cfg_scale}, duration_ms en entero milisegundos, vram_peak_mb (None si no hay CUDA).	true	genera imagen retorna ImageGenResult	python/functions/ml/tests/test_diffusers_backend.py	python/functions/ml/diffusers_generate.py

Ejemplo

from diffusers_load_pipeline import diffusers_load_pipeline
from diffusers_generate import diffusers_generate
from generation_config import GenerationConfig
from model_ref import ModelRef

model = ModelRef(
    name="sd-turbo",
    model_type="sd15",
    path="/home/lucas/vaults/imagegen_models/diffusers/sd-turbo",
)
cfg = GenerationConfig(
    prompt="a photo of a cat",
    seed=42,
    steps=1,
    cfg_scale=0.0,
    sampler="euler",
    width=512,
    height=512,
    model=model,
)
pipe = diffusers_load_pipeline(model, device="cuda", dtype="fp16")
result = diffusers_generate(pipe, cfg)
# result.image -> PIL.Image.Image 512x512
# result.duration_ms -> int > 0
# result.meta["backend"] -> "diffusers"

Notas

cfg.seed = -1 genera seed aleatorio basado en time.time() (reproducible si se guarda en result.meta["seed"]).

VRAM: torch.cuda.reset_peak_memory_stats() antes de inferencia, torch.cuda.max_memory_allocated() // 1024 // 1024 despues.

genconfig_to_diffusers_kwargs omite generator=None; esta funcion lo reemplaza con torch.Generator(device=device).manual_seed(seed).

Import lazy de torch — ImportError descriptivo si no instalado.