pipeKSamplerSDXL v1 (Legacy)¶
Documentation¶
- Class name:
ttN pipeKSamplerSDXL - Category:
🌏 tinyterra/legacy - Output node:
True
This node is designed to manage the sampling process within a specific pipeline, handling the initialization, state management, and output generation for image sampling tasks. It leverages a sampler to initialize states for pipeline and results, conditionally manages image output visibility, and updates the pipeline state based on the sampling results.
Input types¶
Required¶
sdxl_pipe- Represents the pipeline configuration for the sampling process, affecting how images are sampled and processed.
- Comfy dtype:
PIPE_LINE_SDXL - Python dtype:
dict
upscale_method- unknown
- Comfy dtype:
COMBO[STRING] - Python dtype:
unknown
factor- unknown
- Comfy dtype:
FLOAT - Python dtype:
unknown
crop- unknown
- Comfy dtype:
COMBO[STRING] - Python dtype:
unknown
sampler_state- unknown
- Comfy dtype:
COMBO[STRING] - Python dtype:
unknown
base_steps- unknown
- Comfy dtype:
INT - Python dtype:
unknown
refiner_steps- unknown
- Comfy dtype:
INT - Python dtype:
unknown
cfg- unknown
- Comfy dtype:
FLOAT - Python dtype:
unknown
sampler_name- unknown
- Comfy dtype:
COMBO[STRING] - Python dtype:
unknown
scheduler- unknown
- Comfy dtype:
COMBO[STRING] - Python dtype:
unknown
image_output- Controls the visibility and saving behavior of the sampled images, influencing whether images are displayed or saved based on specified conditions.
- Comfy dtype:
COMBO[STRING] - Python dtype:
str
save_prefix- unknown
- Comfy dtype:
STRING - Python dtype:
unknown
Optional¶
seed- unknown
- Comfy dtype:
INT - Python dtype:
unknown
optional_model- unknown
- Comfy dtype:
MODEL - Python dtype:
unknown
optional_positive- unknown
- Comfy dtype:
CONDITIONING - Python dtype:
unknown
optional_negative- unknown
- Comfy dtype:
CONDITIONING - Python dtype:
unknown
optional_vae- unknown
- Comfy dtype:
VAE - Python dtype:
unknown
optional_refiner_model- unknown
- Comfy dtype:
MODEL - Python dtype:
unknown
optional_refiner_positive- unknown
- Comfy dtype:
CONDITIONING - Python dtype:
unknown
optional_refiner_negative- unknown
- Comfy dtype:
CONDITIONING - Python dtype:
unknown
optional_refiner_vae- unknown
- Comfy dtype:
VAE - Python dtype:
unknown
optional_latent- unknown
- Comfy dtype:
LATENT - Python dtype:
unknown
optional_clip- unknown
- Comfy dtype:
CLIP - Python dtype:
unknown
Output types¶
sdxl_pipe- Comfy dtype:
PIPE_LINE_SDXL - unknown
- Python dtype:
unknown
- Comfy dtype:
model- Comfy dtype:
MODEL - unknown
- Python dtype:
unknown
- Comfy dtype:
positive- Comfy dtype:
CONDITIONING - unknown
- Python dtype:
unknown
- Comfy dtype:
negative- Comfy dtype:
CONDITIONING - unknown
- Python dtype:
unknown
- Comfy dtype:
vae- Comfy dtype:
VAE - unknown
- Python dtype:
unknown
- Comfy dtype:
refiner_model- Comfy dtype:
MODEL - unknown
- Python dtype:
unknown
- Comfy dtype:
refiner_positive- Comfy dtype:
CONDITIONING - unknown
- Python dtype:
unknown
- Comfy dtype:
refiner_negative- Comfy dtype:
CONDITIONING - unknown
- Python dtype:
unknown
- Comfy dtype:
refiner_vae- Comfy dtype:
VAE - unknown
- Python dtype:
unknown
- Comfy dtype:
latent- Comfy dtype:
LATENT - unknown
- Python dtype:
unknown
- Comfy dtype:
clip- Comfy dtype:
CLIP - unknown
- Python dtype:
unknown
- Comfy dtype:
image- Comfy dtype:
IMAGE - unknown
- Python dtype:
unknown
- Comfy dtype:
seed- Comfy dtype:
INT - unknown
- Python dtype:
unknown
- Comfy dtype:
ui- Provides a user interface component, typically images, based on the sampling results and the specified image output behavior.
Usage tips¶
- Infra type:
CPU - Common nodes: unknown
Source code¶
class ttN_pipeKSamplerSDXL:
version = '1.0.2'
upscale_methods = ["None", "nearest-exact", "bilinear", "area", "bicubic", "lanczos", "bislerp"]
crop_methods = ["disabled", "center"]
def __init__(self):
pass
@classmethod
def INPUT_TYPES(cls):
return {"required":
{"sdxl_pipe": ("PIPE_LINE_SDXL",),
"upscale_method": (cls.upscale_methods,),
"factor": ("FLOAT", {"default": 2, "min": 0.0, "max": 10.0, "step": 0.25}),
"crop": (cls.crop_methods,),
"sampler_state": (["Sample", "Hold"], ),
"base_steps": ("INT", {"default": 20, "min": 1, "max": 10000}),
"refiner_steps": ("INT", {"default": 20, "min": 0, "max": 10000}),
"cfg": ("FLOAT", {"default": 8.0, "min": 0.0, "max": 100.0}),
"sampler_name": (comfy.samplers.KSampler.SAMPLERS,),
"scheduler": (comfy.samplers.KSampler.SCHEDULERS,),
"image_output": (["Hide", "Preview", "Save", "Hide/Save"],),
"save_prefix": ("STRING", {"default": "ComfyUI"})
},
"optional":
{"seed": ("INT", {"default": 0, "min": 0, "max": 0xffffffffffffffff}),
"optional_model": ("MODEL",),
"optional_positive": ("CONDITIONING",),
"optional_negative": ("CONDITIONING",),
"optional_vae": ("VAE",),
"optional_refiner_model": ("MODEL",),
"optional_refiner_positive": ("CONDITIONING",),
"optional_refiner_negative": ("CONDITIONING",),
"optional_refiner_vae": ("VAE",),
"optional_latent": ("LATENT",),
"optional_clip": ("CLIP",),
#"xyPlot": ("XYPLOT",),
},
"hidden":
{"prompt": "PROMPT", "extra_pnginfo": "EXTRA_PNGINFO", "my_unique_id": "UNIQUE_ID",
"embeddingsList": (folder_paths.get_filename_list("embeddings"),),
"ttNnodeVersion": ttN_pipeKSamplerSDXL.version
},
}
RETURN_TYPES = ("PIPE_LINE_SDXL", "MODEL", "CONDITIONING", "CONDITIONING", "VAE", "MODEL", "CONDITIONING", "CONDITIONING", "VAE", "LATENT", "CLIP", "IMAGE", "INT",)
RETURN_NAMES = ("sdxl_pipe", "model", "positive", "negative" ,"vae", "refiner_model", "refiner_positive", "refiner_negative" ,"refiner_vae", "latent", "clip", "image", "seed", )
OUTPUT_NODE = True
FUNCTION = "sample"
CATEGORY = "🌏 tinyterra/legacy"
def sample(self, sdxl_pipe, sampler_state,
base_steps, refiner_steps, cfg, sampler_name, scheduler, image_output, save_prefix, denoise=1.0,
optional_model=None, optional_positive=None, optional_negative=None, optional_latent=None, optional_vae=None, optional_clip=None,
optional_refiner_model=None, optional_refiner_positive=None, optional_refiner_negative=None, optional_refiner_vae=None,
seed=None, xyPlot=None, upscale_method=None, factor=None, crop=None, prompt=None, extra_pnginfo=None, my_unique_id=None,
start_step=None, last_step=None, force_full_denoise=False, disable_noise=False):
sdxl_pipe = {**sdxl_pipe}
my_unique_id = int(my_unique_id)
ttN_save = ttNsave(my_unique_id, prompt, extra_pnginfo)
sdxl_samples = optional_latent if optional_latent is not None else sdxl_pipe["samples"]
sdxl_model = optional_model if optional_model is not None else sdxl_pipe["model"]
sdxl_positive = optional_positive if optional_positive is not None else sdxl_pipe["positive"]
sdxl_negative = optional_negative if optional_negative is not None else sdxl_pipe["negative"]
sdxl_vae = optional_vae if optional_vae is not None else sdxl_pipe["vae"]
sdxl_clip = optional_clip if optional_clip is not None else sdxl_pipe["clip"]
sdxl_refiner_model = optional_refiner_model if optional_refiner_model is not None else sdxl_pipe["refiner_model"]
sdxl_refiner_positive = optional_refiner_positive if optional_refiner_positive is not None else sdxl_pipe["refiner_positive"]
sdxl_refiner_negative = optional_refiner_negative if optional_refiner_negative is not None else sdxl_pipe["refiner_negative"]
sdxl_refiner_vae = optional_refiner_vae if optional_refiner_vae is not None else sdxl_pipe["refiner_vae"]
sdxl_refiner_clip = sdxl_pipe["refiner_clip"]
if seed in (None, 'undefined'):
sdxl_seed = sdxl_pipe["seed"]
else:
sdxl_seed = seed
def process_sample_state(sdxl_pipe, sdxl_samples, sdxl_model, sdxl_positive, sdxl_negative, sdxl_vae, sdxl_clip, sdxl_seed,
sdxl_refiner_model, sdxl_refiner_positive, sdxl_refiner_negative, sdxl_refiner_vae, sdxl_refiner_clip,
base_steps, refiner_steps, cfg, sampler_name, scheduler, denoise,
image_output, save_prefix, prompt, my_unique_id, preview_latent, disable_noise=disable_noise):
total_steps = base_steps + refiner_steps
# Upscale samples if enabled
sdxl_samples = sampler.handle_upscale(sdxl_samples, upscale_method, factor, crop)
if (refiner_steps > 0) and (sdxl_refiner_model not in [None, "None"]):
# Base Sample
sdxl_samples = sampler.common_ksampler(sdxl_model, sdxl_seed, total_steps, cfg, sampler_name, scheduler, sdxl_positive, sdxl_negative, sdxl_samples,
denoise=denoise, preview_latent=preview_latent, start_step=0, last_step=base_steps, force_full_denoise=force_full_denoise, disable_noise=disable_noise)
# Refiner Sample
sdxl_samples = sampler.common_ksampler(sdxl_refiner_model, sdxl_seed, total_steps, cfg, sampler_name, scheduler, sdxl_refiner_positive, sdxl_refiner_negative, sdxl_samples,
denoise=denoise, preview_latent=preview_latent, start_step=base_steps, last_step=10000, force_full_denoise=True, disable_noise=True)
latent = sdxl_samples["samples"]
sdxl_images = sdxl_refiner_vae.decode(latent)
del latent
else:
sdxl_samples = sampler.common_ksampler(sdxl_model, sdxl_seed, base_steps, cfg, sampler_name, scheduler, sdxl_positive, sdxl_negative, sdxl_samples,
denoise=denoise, preview_latent=preview_latent, start_step=0, last_step=base_steps, force_full_denoise=True, disable_noise=disable_noise)
latent = sdxl_samples["samples"]
sdxl_images = sdxl_vae.decode(latent)
del latent
results = ttN_save.images(sdxl_images, save_prefix, image_output)
sampler.update_value_by_id("results", my_unique_id, results)
new_sdxl_pipe = {"model": sdxl_model,
"positive": sdxl_positive,
"negative": sdxl_negative,
"vae": sdxl_vae,
"clip": sdxl_clip,
"refiner_model": sdxl_refiner_model,
"refiner_positive": sdxl_refiner_positive,
"refiner_negative": sdxl_refiner_negative,
"refiner_vae": sdxl_refiner_vae,
"refiner_clip": sdxl_refiner_clip,
"samples": sdxl_samples,
"images": sdxl_images,
"seed": sdxl_seed,
"loader_settings": sdxl_pipe["loader_settings"],
}
del sdxl_pipe
sampler.update_value_by_id("pipe_line", my_unique_id, new_sdxl_pipe)
if image_output in ("Hide", "Hide/Save"):
return sampler.get_output_sdxl(new_sdxl_pipe)
return {"ui": {"images": results},
"result": sampler.get_output_sdxl(new_sdxl_pipe)}
def process_hold_state(sdxl_pipe, image_output, my_unique_id):
ttNl('Held').t(f'pipeKSamplerSDXL[{my_unique_id}]').p()
last_pipe = sampler.init_state(my_unique_id, "pipe_line", sdxl_pipe)
last_results = sampler.init_state(my_unique_id, "results", list())
if image_output in ("Hide", "Hide/Save"):
return sampler.get_output_sdxl(last_pipe)
return {"ui": {"images": last_results}, "result": sampler.get_output_sdxl(last_pipe)}
preview_latent = True
if image_output in ("Hide", "Hide/Save"):
preview_latent = False
if sampler_state == "Sample" and xyPlot is None:
return process_sample_state(sdxl_pipe, sdxl_samples, sdxl_model, sdxl_positive, sdxl_negative, sdxl_vae, sdxl_clip, sdxl_seed,
sdxl_refiner_model, sdxl_refiner_positive, sdxl_refiner_negative, sdxl_refiner_vae, sdxl_refiner_clip, base_steps, refiner_steps, cfg, sampler_name, scheduler, denoise, image_output, save_prefix, prompt, my_unique_id, preview_latent)
#elif sampler_state == "Sample" and xyPlot is not None:
# return process_xyPlot(sdxl_pipe, lora_name, lora_model_strength, lora_clip_strength, steps, cfg, sampler_name, scheduler, denoise, image_output, save_prefix, prompt, extra_pnginfo, my_unique_id, preview_latent, xyPlot)
elif sampler_state == "Hold":
return process_hold_state(sdxl_pipe, image_output, my_unique_id)