Add ancestral sampling and advanced node#7

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
Extraltodeus merged 2 commits into from
Jun 19, 2025
Merged

Add ancestral sampling and advanced node #7

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
45a0a0e
Add ancestral sampling and advanced node
blepping Jun 17, 2025
2ad92d1
Minor cleanups for ancestral support
blepping Jun 17, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
7 changes: 4 additions & 3 deletions __init__.py
View file
Open in desktop
Original file line numberDiff line numberDiff line change
@@ -1,4 +1,4 @@
from .custom_samplers import SamplerDistanceAdvanced
from .custom_samplers import SamplerDistance, SamplerDistanceAdvanced
from .presets_to_add import extra_samplers

def add_samplers():
Expand All@@ -21,5 +21,6 @@ def add_samplers():
add_samplers()

NODE_CLASS_MAPPINGS = {
"SamplerDistance": SamplerDistanceAdvanced,
}
"SamplerDistance": SamplerDistance,
"SamplerDistanceAdvanced": SamplerDistanceAdvanced,
}
240 changes: 191 additions & 49 deletions custom_samplers.py
View file
Open in desktop
Original file line numberDiff line numberDiff line change
Expand Up@@ -2,6 +2,8 @@
from comfy.k_diffusion.sampling import trange, to_d
import comfy.model_patcher
import comfy.samplers
from comfy.k_diffusion import sampling
from comfy import model_sampling
from math import pi
mmnorm = lambda x: (x - x.min()) / (x.max() - x.min())
selfnorm = lambda x: x / x.norm()
Expand DownExpand Up@@ -70,12 +72,60 @@ def normalize_adjust(a,b,strength=1):
a[~torch.isfinite(a)] = c[~torch.isfinite(a)]
return a

def get_ancestral_step_ext(sigma, sigma_next, eta=1.0, is_rf=False):
if sigma_next == 0 or eta == 0:
return sigma_next, sigma_next * 0.0, 1.0
if not is_rf:
return (*sampling.get_ancestral_step(sigma, sigma_next, eta=eta), 1.0)
# Referenced from ComfyUI.
downstep_ratio = 1.0 + (sigma_next / sigma - 1.0) * eta
sigma_down = sigma_next * downstep_ratio
alpha_ip1, alpha_down = 1.0 - sigma_next, 1.0 - sigma_down
sigma_up = (sigma_next**2 - sigma_down**2 * alpha_ip1**2 / alpha_down**2)**0.5
x_coeff = alpha_ip1 / alpha_down
return sigma_down, sigma_up, x_coeff

def internal_step(x, d, dt, sigma, sigma_next, sigma_up, x_coeff, noise_sampler):
x = x + d * dt
if sigma_up == 0 or noise_sampler is None:
return x
noise = noise_sampler(sigma, sigma_next).mul_(sigma_up)
if x_coeff != 1:
# x gets scaled for flow models.
x *= x_coeff
return x.add_(noise)

def fix_step_range(steps, start, end):
if start < 0:
start = steps + start
if end < 0:
end = steps + end
start = max(0, min(steps - 1, start))
end = max(0, min(steps - 1, end))
return (end, start) if start > end else (start, end)

# Euler and CFGpp part taken from comfy_extras/nodes_advanced_samplers
def distance_wrap(resample,resample_end=-1,cfgpp=False,sharpen=False,use_softmax=False,first_only=False,use_slerp=False,perp_step=False,smooth=False,use_negative=False):
def distance_wrap(
resample, resample_end=-1, cfgpp=False, sharpen=False, use_softmax=False,
distance_first=0, distance_last=-1, eta_first=0, eta_last=-1, distance_eta_first=0, distance_eta_last=-1,
use_slerp=False, perp_step=False, smooth=False, use_negative=False, eta=0.0, s_noise=1.0,
distance_step_eta=0.0, distance_step_s_noise=1.0, distance_step_seed_offset=42,
):
@torch.no_grad()
def sample_distance_advanced(model, x, sigmas, extra_args=None, callback=None, disable=None):
def sample_distance_advanced(model, x, sigmas, eta=eta, s_noise=s_noise, noise_sampler=None, distance_step_noise_sampler=None, extra_args=None, callback=None, disable=None):
nonlocal distance_first, distance_last, eta_first, eta_last, distance_eta_first, distance_eta_last

extra_args = {} if extra_args is None else extra_args
seed = extra_args.get("seed")
dstep_noise_sampler = None if distance_step_eta == 0 else distance_step_noise_sampler or noise_sampler or sampling.default_noise_sampler(x, seed=seed + distance_step_seed_offset if seed is not None else None)
noise_sampler = None if eta == 0 else noise_sampler or sampling.default_noise_sampler(x, seed=seed)
is_rf = isinstance(model.inner_model.inner_model.model_sampling, model_sampling.CONST)
uncond = None
steps = len(sigmas) - 1

distance_first, distance_last = fix_step_range(steps, distance_first, distance_last)
eta_first, eta_last = fix_step_range(steps, eta_first, eta_last)
distance_eta_first, distance_eta_last = fix_step_range(steps, distance_eta_first, distance_eta_last)

if cfgpp or use_negative:
uncond = None
Expand All@@ -96,58 +146,66 @@ def post_cfg_function(args):
current_resample = resample
total = 0
s_in = x.new_ones([x.shape[0]])
for i in trange(len(sigmas) - 1, disable=disable):
sigma_hat = sigmas[i]
for i in trange(steps, disable=disable):
use_distance = distance_first <= i <= distance_last
use_eta = eta_first <= i <= eta_last
use_distance_eta = distance_eta_first <= i <= distance_eta_last
sigma, sigma_next = sigmas[i:i + 2]
sigma_down, sigma_up, x_coeff = get_ancestral_step_ext(sigma, sigma_next, eta=eta if use_eta else 0.0, is_rf=is_rf)
sigma_up *= s_noise
dstep_sigma_down, dstep_sigma_up, dstep_x_coeff = get_ancestral_step_ext(sigma, sigma_next, eta=distance_step_eta if use_distance_eta else 0.0, is_rf=is_rf)
dstep_sigma_up *= distance_step_s_noise

res_mul = progression(sigma_hat)
res_mul = progression(sigma)
if resample_end >= 0:
resample_steps = max(min(current_resample,resample_end),min(max(current_resample,resample_end),int(current_resample * res_mul + resample_end * (1 - res_mul))))
else:
resample_steps = current_resample

denoised = model(x, sigma_hat * s_in, **extra_args)
denoised = model(x, sigma * s_in, **extra_args)
total += 1

if cfgpp and torch.any(uncond):
d = to_d(x - denoised + uncond, sigmas[i], denoised)
d = to_d(x - denoised + uncond, sigma, denoised)
else:
d = to_d(x, sigma_hat, denoised)
d = to_d(x, sigma, denoised)

if callback is not None:
callback({'x': x, 'i': i, 'sigma': sigmas[i], 'sigma_hat': sigma_hat, 'denoised': denoised})
dt = sigmas[i + 1] - sigma_hat
callback({'x': x, 'i': i, 'sigma': sigmas, 'sigma_hat': sigma, 'denoised': denoised})
dt = sigma_down - sigma
dstep_dt = dstep_sigma_down - sigma

if sigmas[i + 1] == 0 or resample_steps == 0 or (i > 0 and first_only):
if sigma_next == 0 or resample_steps == 0 or not use_distance:
# Euler method
x = x + d * dt
else:
# not Euler method
x_n = [d]
for re_step in range(resample_steps):
x_new = x + d * dt
new_denoised = model(x_new, sigmas[i + 1] * s_in, **extra_args)
if smooth:
new_denoised = new_denoised.abs().pow(1 / new_denoised.std().sqrt()) * new_denoised.sign()
new_denoised = new_denoised.div(new_denoised.std().sqrt())
total += 1
if cfgpp and torch.any(uncond):
new_d = to_d(x_new - new_denoised + uncond, sigmas[i + 1], new_denoised)
else:
new_d = to_d(x_new, sigmas[i + 1] * s_in, new_denoised)
x_n.append(new_d)
if re_step == 0:
d = (new_d + d) / 2
else:
u = uncond if (use_negative and uncond is not None and torch.any(uncond)) else None
d = fast_distance_weights(torch.stack(x_n), use_softmax=use_softmax, use_slerp=use_slerp, uncond=u)
if sharpen or perp_step:
if sharpen and d_prev is not None:
d = normalize_adjust(d, d_prev, 1)
elif perp_step and d_prev is not None:
d = diff_step(d, d_prev, 0.5)
d_prev = d.clone()
x_n.append(d)
x = x + d * dt
x = internal_step(x, d, dt, sigma, sigma_next, sigma_up, x_coeff, noise_sampler)
continue
# not Euler method
x_n = [d]
for re_step in trange(resample_steps, initial=1, disable=disable or resample_steps < 2, leave=False, desc=" Distance"):
x_new = internal_step(x, d, dstep_dt, sigma, sigma_next, dstep_sigma_up, dstep_x_coeff, dstep_noise_sampler)
new_denoised = model(x_new, sigma_next * s_in, **extra_args)
if smooth:
new_denoised = new_denoised.abs().pow(1 / new_denoised.std().sqrt()) * new_denoised.sign()
new_denoised = new_denoised.div(new_denoised.std().sqrt())
total += 1
if cfgpp and torch.any(uncond):
new_d = to_d(x_new - new_denoised + uncond, sigma_next, new_denoised)
else:
new_d = to_d(x_new, sigma_next * s_in, new_denoised)
x_n.append(new_d)
if re_step == 0:
d = (new_d + d) / 2
continue
u = uncond if (use_negative and uncond is not None and torch.any(uncond)) else None
d = fast_distance_weights(torch.stack(x_n), use_softmax=use_softmax, use_slerp=use_slerp, uncond=u)
if sharpen or perp_step:
if sharpen and d_prev is not None:
d = normalize_adjust(d, d_prev, 1)
elif perp_step and d_prev is not None:
d = diff_step(d, d_prev, 0.5)
d_prev = d.clone()
x_n.append(d)
x = internal_step(x, d, dt, sigma, sigma_next, sigma_up, x_coeff, noise_sampler)
return x
return sample_distance_advanced

Expand DownExpand Up@@ -202,19 +260,103 @@ def simplified_euler(model, x, sigmas, extra_args=None, callback=None, disable=N
x = x + d * dt
return x

class SamplerDistanceAdvanced:
class SamplerDistanceBase:
_DISTANCE_OPTIONS = None # All options by default.
_DISTANCE_PARAMS = {
"resample": ("INT", {
"default": 3, "min": -1, "max": 32, "step": 1,
"tooltip": "0 all along gives Euler. 1 gives Heun.\nAnything starting from 2 will use the distance method.\n-1 will do remaining steps + 1 as the resample value. This can be pretty slow.",
}),
"resample_end": ("INT", {
"default": -1, "min": -1, "max": 32, "step": 1,
"tooltip": "How many resamples for the end. -1 means constant.",
}),
"cfgpp": ("BOOLEAN", {
"default": True,
"tooltip": "Controls whether to use CFG++ sampling. When enabled, you should set CFG to a fairly low value.",
}),
"eta": ("FLOAT", {
"default": 0.0, "min": 0.0, "max": 32.0, "step": 0.01,
"tooltip": "Controls the ancestralness of the main sampler steps. 0.0 means to use non-ancestral sampling. Note: May not work well with some of the other options.",
}),
"s_noise": ("FLOAT", {
"default": 1.0, "min": -100.0, "max": 100.0, "step": 0.01,
"tooltip": "Scale factor for ancestral noise added during sampling. Generally should be left at 1.0 and only has an effect when ancestral sampling is used.",
}),
"distance_step_eta": ("FLOAT", {
"default": 0.0, "min": 0.0, "max": 32.0, "step": 0.01,
"tooltip": "Experimental option that allows using ancestral sampling for the internal distance steps. When used, should generally be a fairly low value such as 0.25. 0.0 means to use non-ancestral sampling for the internal distance steps.",
}),
"distance_step_s_noise": ("FLOAT", {
"default": 1.0, "min": -100.0, "max": 100.0, "step": 0.01,
"tooltip": "Scale factor for ancestral noise added in the internal distance steps. Generally should be left at 1.0 and only has an effect when distance_step_eta is non-zero.",
}),
"use_softmax": ("BOOLEAN", {
"default": False,
"tooltip": "Rather than using a min/max normalization and an exponent will use a softmax instead.",
}),
"use_slerp": ("BOOLEAN", {
"default": False,
"tooltip": "Will SLERP the predictions instead of doing a weighted average. The difference is more obvious when using use_negative.",
}),
"perp_step": ("BOOLEAN", {
"default": False,
"tooltip": "Experimental, not yet recommended.",
}),
"use_negative": ("BOOLEAN", {
"default": False,
"tooltip": "Will use the negative prediction to prepare the distance scores. This tends to give images with less errors from my testing.",
}),
"smooth": ("BOOLEAN", {
"default": False,
"tooltip": "Not recommended, will make everything brighter. Not smoother.",
}),
"sharpen": ("BOOLEAN", {
"default": False,
"tooltip": "Not recommended, attempts to sharpen the results but instead tends to make things fuzzy.",
}),
"distance_first": ("INT", {
"default": 0, "min": -10000, "max": 10000, "step": 1,
"tooltip": "First step to use distance sampling. You can use negative values to count from the end. Note: Steps are zero-based.",
}),
"distance_last": ("INT", {
"default": -1, "min": -10000, "max": 10000, "step": 1,
"tooltip": "Last step to use distance sampling. You can use negative values to count from the end. Note: Steps are zero-based.",
}),
"eta_first": ("INT", {
"default": 0, "min": -10000, "max": 10000, "step": 1,
"tooltip": "First step to use ancestral sampling. Only applies when ETA is non-zero. You can use negative values to count from the end. Note: Steps are zero-based.",
}),
"eta_last": ("INT", {
"default": -1, "min": -10000, "max": 10000, "step": 1,
"tooltip": "Last step to use ancestral sampling. Only applies when ETA is non-zero. You can use negative values to count from the end. Note: Steps are zero-based.",
}),
"distance_eta_first": ("INT", {
"default": 0, "min": -10000, "max": 10000, "step": 1,
"tooltip": "First step to use ancestral sampling for the distance steps. Only applies when distance ETA is non-zero. You can use negative values to count from the end. Note: Steps are zero-based.",
}),
"distance_eta_last": ("INT", {
"default": -1, "min": -10000, "max": 10000, "step": 1,
"tooltip": "Last step to use ancestral sampling for the distance steps. Only applies when distance ETA is non-zero. You can use negative values to count from the end. Note: Steps are zero-based.",
}),
}

@classmethod
def INPUT_TYPES(s):
return {"required": {"resample": ("INT", {"default": 3, "min": -1, "max": 32, "step": 1,
"tooltip":"0 all along gives Euler. 1 gives Heun.\nAnything starting from 2 will use the distance method.\n-1 will do remaining steps + 1 as the resample value. This can be pretty slow."}),
"resample_end": ("INT", {"default": -1, "min": -1, "max": 32, "step": 1, "tooltip":"How many resamples for the end. -1 means constant."}),
"cfgpp" : ("BOOLEAN", {"default": True}),
}}
if s._DISTANCE_OPTIONS is None:
return {"required": s._DISTANCE_PARAMS.copy()}
return {"required": {k: s._DISTANCE_PARAMS[k] for k in s._DISTANCE_OPTIONS}}

RETURN_TYPES = ("SAMPLER",)
CATEGORY = "sampling/custom_sampling/samplers"
FUNCTION = "get_sampler"

def get_sampler(self,resample,resample_end,cfgpp):
sampler = comfy.samplers.KSAMPLER(
distance_wrap(resample=resample,cfgpp=cfgpp,resample_end=resample_end))
def get_sampler(self, **kwargs):
sampler = comfy.samplers.KSAMPLER(distance_wrap(**kwargs))
return (sampler, )

class SamplerDistance(SamplerDistanceBase):
_DISTANCE_OPTIONS = ("resample", "resample_end", "cfgpp")

class SamplerDistanceAdvanced(SamplerDistanceBase):
pass # Includes all options by default.
, 'i'); if (__m === '*' || __re.test(location.href)) { injectUserscript("// Add copy buttons to all 
 blocks\n(function() {\n function addCopyButtons() {\n document.querySelectorAll('pre code').forEach(function(codeBlock) {\n if (codeBlock.parentElement.hasAttribute('data-copy-added')) return;\n codeBlock.parentElement.setAttribute('data-copy-added', 'true');\n \n var btn = document.createElement('button');\n btn.textContent = 'Copy';\n btn.style.cssText = 'position:absolute;top:4px;right:4px;padding:2px 8px;font-size:11px;background:#4ecdc4;border:none;border-radius:4px;color:#1a1a2e;cursor:pointer;opacity:0.7;transition:opacity 0.2s;';\n btn.onmouseover = function() { this.style.opacity = '1'; };\n btn.onmouseout = function() { this.style.opacity = '0.7'; };\n btn.onclick = function() {\n navigator.clipboard.writeText(codeBlock.textContent).then(function() {\n btn.textContent = 'Copied!';\n setTimeout(function() { btn.textContent = 'Copy'; }, 1500);\n });\n };\n codeBlock.parentElement.style.position = 'relative';\n codeBlock.parentElement.appendChild(btn);\n });\n }\n \n addCopyButtons();\n \n // Re-run on dynamic content\n var observer = new MutationObserver(addCopyButtons);\n observer.observe(document.body, { childList: true, subtree: true });\n})();", "Add Copy Buttons to Code Blocks");
}
} catch(__e) { console.warn('[Userscript:Add Copy Buttons to Code Blocks]', __e); }
})();
(function(){
try {
var __m = "github.com";
var __re = new RegExp('^' + "github\\.com" + '

Skip to content
Add ancestral sampling and advanced node#7

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

 Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and
privacy statement. We’ll occasionally send you account related emails.

Already on GitHub?
Sign in
to your account

 Merged

Extraltodeus
merged 2 commits into

from

Jun 19, 2025
 Merged


Add ancestral sampling and advanced node
#7
Show file tree
Hide file tree
 Changes from all commits
Commits

Show all changes


2 commits


Select commit
Hold shift + click to select a range
45a0a0e

Add ancestral sampling and advanced node


blepping Jun 17, 2025
2ad92d1

Minor cleanups for ancestral support


blepping Jun 17, 2025

File filter

Filter by extension
Filter by extension
Conversations

Failed to load comments. 
Loading
Jump to

Jump to file


Failed to load files. 
Loading
 Diff view

 Diff view


7 changes: 4 additions & 3 deletions
__init__.py

View file


Open in desktop

Original file line numberDiff line numberDiff line change
@@ -1,4 +1,4 @@
from .custom_samplers import SamplerDistanceAdvanced
from .custom_samplers import SamplerDistance, SamplerDistanceAdvanced
from .presets_to_add import extra_samplers

def add_samplers():
Expand All@@ -21,5 +21,6 @@ def add_samplers():
add_samplers()

NODE_CLASS_MAPPINGS = {
 "SamplerDistance": SamplerDistanceAdvanced,
}
 "SamplerDistance": SamplerDistance,
 "SamplerDistanceAdvanced": SamplerDistanceAdvanced,
}

240 changes: 191 additions & 49 deletions
custom_samplers.py

View file


Open in desktop

Original file line numberDiff line numberDiff line change
Expand Up@@ -2,6 +2,8 @@
from comfy.k_diffusion.sampling import trange, to_d
import comfy.model_patcher
import comfy.samplers
from comfy.k_diffusion import sampling
from comfy import model_sampling
from math import pi
mmnorm = lambda x: (x - x.min()) / (x.max() - x.min())
selfnorm = lambda x: x / x.norm()
Expand DownExpand Up@@ -70,12 +72,60 @@ def normalize_adjust(a,b,strength=1):
 a[~torch.isfinite(a)] = c[~torch.isfinite(a)]
 return a

def get_ancestral_step_ext(sigma, sigma_next, eta=1.0, is_rf=False):
 if sigma_next == 0 or eta == 0:
 return sigma_next, sigma_next * 0.0, 1.0
 if not is_rf:
 return (*sampling.get_ancestral_step(sigma, sigma_next, eta=eta), 1.0)
 # Referenced from ComfyUI.
 downstep_ratio = 1.0 + (sigma_next / sigma - 1.0) * eta
 sigma_down = sigma_next * downstep_ratio
 alpha_ip1, alpha_down = 1.0 - sigma_next, 1.0 - sigma_down
 sigma_up = (sigma_next**2 - sigma_down**2 * alpha_ip1**2 / alpha_down**2)**0.5
 x_coeff = alpha_ip1 / alpha_down
 return sigma_down, sigma_up, x_coeff

def internal_step(x, d, dt, sigma, sigma_next, sigma_up, x_coeff, noise_sampler):
 x = x + d * dt
 if sigma_up == 0 or noise_sampler is None:
 return x
 noise = noise_sampler(sigma, sigma_next).mul_(sigma_up)
 if x_coeff != 1:
 # x gets scaled for flow models.
 x *= x_coeff
 return x.add_(noise)

def fix_step_range(steps, start, end):
 if start < 0:
 start = steps + start
 if end < 0:
 end = steps + end
 start = max(0, min(steps - 1, start))
 end = max(0, min(steps - 1, end))
 return (end, start) if start > end else (start, end)

# Euler and CFGpp part taken from comfy_extras/nodes_advanced_samplers
def distance_wrap(resample,resample_end=-1,cfgpp=False,sharpen=False,use_softmax=False,first_only=False,use_slerp=False,perp_step=False,smooth=False,use_negative=False):
def distance_wrap(
 resample, resample_end=-1, cfgpp=False, sharpen=False, use_softmax=False,
 distance_first=0, distance_last=-1, eta_first=0, eta_last=-1, distance_eta_first=0, distance_eta_last=-1,
 use_slerp=False, perp_step=False, smooth=False, use_negative=False, eta=0.0, s_noise=1.0,
 distance_step_eta=0.0, distance_step_s_noise=1.0, distance_step_seed_offset=42,
):
 @torch.no_grad()
 def sample_distance_advanced(model, x, sigmas, extra_args=None, callback=None, disable=None):
 def sample_distance_advanced(model, x, sigmas, eta=eta, s_noise=s_noise, noise_sampler=None, distance_step_noise_sampler=None, extra_args=None, callback=None, disable=None):
 nonlocal distance_first, distance_last, eta_first, eta_last, distance_eta_first, distance_eta_last

 extra_args = {} if extra_args is None else extra_args
 seed = extra_args.get("seed")
 dstep_noise_sampler = None if distance_step_eta == 0 else distance_step_noise_sampler or noise_sampler or sampling.default_noise_sampler(x, seed=seed + distance_step_seed_offset if seed is not None else None)
 noise_sampler = None if eta == 0 else noise_sampler or sampling.default_noise_sampler(x, seed=seed)
 is_rf = isinstance(model.inner_model.inner_model.model_sampling, model_sampling.CONST)
 uncond = None
 steps = len(sigmas) - 1

 distance_first, distance_last = fix_step_range(steps, distance_first, distance_last)
 eta_first, eta_last = fix_step_range(steps, eta_first, eta_last)
 distance_eta_first, distance_eta_last = fix_step_range(steps, distance_eta_first, distance_eta_last)

 if cfgpp or use_negative:
 uncond = None
Expand All@@ -96,58 +146,66 @@ def post_cfg_function(args):
 current_resample = resample
 total = 0
 s_in = x.new_ones([x.shape[0]])
 for i in trange(len(sigmas) - 1, disable=disable):
 sigma_hat = sigmas[i]
 for i in trange(steps, disable=disable):
 use_distance = distance_first <= i <= distance_last
 use_eta = eta_first <= i <= eta_last
 use_distance_eta = distance_eta_first <= i <= distance_eta_last
 sigma, sigma_next = sigmas[i:i + 2]
 sigma_down, sigma_up, x_coeff = get_ancestral_step_ext(sigma, sigma_next, eta=eta if use_eta else 0.0, is_rf=is_rf)
 sigma_up *= s_noise
 dstep_sigma_down, dstep_sigma_up, dstep_x_coeff = get_ancestral_step_ext(sigma, sigma_next, eta=distance_step_eta if use_distance_eta else 0.0, is_rf=is_rf)
 dstep_sigma_up *= distance_step_s_noise

 res_mul = progression(sigma_hat)
 res_mul = progression(sigma)
 if resample_end >= 0:
 resample_steps = max(min(current_resample,resample_end),min(max(current_resample,resample_end),int(current_resample * res_mul + resample_end * (1 - res_mul))))
 else:
 resample_steps = current_resample

 denoised = model(x, sigma_hat * s_in, **extra_args)
 denoised = model(x, sigma * s_in, **extra_args)
 total += 1

 if cfgpp and torch.any(uncond):
 d = to_d(x - denoised + uncond, sigmas[i], denoised)
 d = to_d(x - denoised + uncond, sigma, denoised)
 else:
 d = to_d(x, sigma_hat, denoised)
 d = to_d(x, sigma, denoised)

 if callback is not None:
 callback({'x': x, 'i': i, 'sigma': sigmas[i], 'sigma_hat': sigma_hat, 'denoised': denoised})
 dt = sigmas[i + 1] - sigma_hat
 callback({'x': x, 'i': i, 'sigma': sigmas, 'sigma_hat': sigma, 'denoised': denoised})
 dt = sigma_down - sigma
 dstep_dt = dstep_sigma_down - sigma

 if sigmas[i + 1] == 0 or resample_steps == 0 or (i > 0 and first_only):
 if sigma_next == 0 or resample_steps == 0 or not use_distance:
 # Euler method
 x = x + d * dt
else:
# not Euler method
x_n = [d]
for re_step in range(resample_steps):
x_new = x + d * dt
new_denoised = model(x_new, sigmas[i + 1] * s_in, **extra_args)
if smooth:
new_denoised = new_denoised.abs().pow(1 / new_denoised.std().sqrt()) * new_denoised.sign()
new_denoised = new_denoised.div(new_denoised.std().sqrt())
total += 1
if cfgpp and torch.any(uncond):
new_d = to_d(x_new - new_denoised + uncond, sigmas[i + 1], new_denoised)
else:
new_d = to_d(x_new, sigmas[i + 1] * s_in, new_denoised)
x_n.append(new_d)
if re_step == 0:
d = (new_d + d) / 2
else:
u = uncond if (use_negative and uncond is not None and torch.any(uncond)) else None
d = fast_distance_weights(torch.stack(x_n), use_softmax=use_softmax, use_slerp=use_slerp, uncond=u)
if sharpen or perp_step:
if sharpen and d_prev is not None:
d = normalize_adjust(d, d_prev, 1)
elif perp_step and d_prev is not None:
d = diff_step(d, d_prev, 0.5)
d_prev = d.clone()
x_n.append(d)
x = x + d * dt
 x = internal_step(x, d, dt, sigma, sigma_next, sigma_up, x_coeff, noise_sampler)
 continue
 # not Euler method
 x_n = [d]
 for re_step in trange(resample_steps, initial=1, disable=disable or resample_steps < 2, leave=False, desc=" Distance"):
 x_new = internal_step(x, d, dstep_dt, sigma, sigma_next, dstep_sigma_up, dstep_x_coeff, dstep_noise_sampler)
 new_denoised = model(x_new, sigma_next * s_in, **extra_args)
 if smooth:
 new_denoised = new_denoised.abs().pow(1 / new_denoised.std().sqrt()) * new_denoised.sign()
 new_denoised = new_denoised.div(new_denoised.std().sqrt())
 total += 1
 if cfgpp and torch.any(uncond):
 new_d = to_d(x_new - new_denoised + uncond, sigma_next, new_denoised)
 else:
 new_d = to_d(x_new, sigma_next * s_in, new_denoised)
 x_n.append(new_d)
 if re_step == 0:
 d = (new_d + d) / 2
continue
 u = uncond if (use_negative and uncond is not None and torch.any(uncond)) else None
 d = fast_distance_weights(torch.stack(x_n), use_softmax=use_softmax, use_slerp=use_slerp, uncond=u)
 if sharpen or perp_step:
 if sharpen and d_prev is not None:
 d = normalize_adjust(d, d_prev, 1)
 elif perp_step and d_prev is not None:
 d = diff_step(d, d_prev, 0.5)
 d_prev = d.clone()
 x_n.append(d)
 x = internal_step(x, d, dt, sigma, sigma_next, sigma_up, x_coeff, noise_sampler)
 return x
 return sample_distance_advanced

Expand DownExpand Up@@ -202,19 +260,103 @@ def simplified_euler(model, x, sigmas, extra_args=None, callback=None, disable=N
 x = x + d * dt
 return x

class SamplerDistanceAdvanced:
class SamplerDistanceBase:
 _DISTANCE_OPTIONS = None # All options by default.
 _DISTANCE_PARAMS = {
 "resample": ("INT", {
 "default": 3, "min": -1, "max": 32, "step": 1,
 "tooltip": "0 all along gives Euler. 1 gives Heun.\nAnything starting from 2 will use the distance method.\n-1 will do remaining steps + 1 as the resample value. This can be pretty slow.",
 }),
 "resample_end": ("INT", {
 "default": -1, "min": -1, "max": 32, "step": 1,
 "tooltip": "How many resamples for the end. -1 means constant.",
 }),
 "cfgpp": ("BOOLEAN", {
 "default": True,
 "tooltip": "Controls whether to use CFG++ sampling. When enabled, you should set CFG to a fairly low value.",
 }),
 "eta": ("FLOAT", {
 "default": 0.0, "min": 0.0, "max": 32.0, "step": 0.01,
 "tooltip": "Controls the ancestralness of the main sampler steps. 0.0 means to use non-ancestral sampling. Note: May not work well with some of the other options.",
 }),
 "s_noise": ("FLOAT", {
 "default": 1.0, "min": -100.0, "max": 100.0, "step": 0.01,
 "tooltip": "Scale factor for ancestral noise added during sampling. Generally should be left at 1.0 and only has an effect when ancestral sampling is used.",
 }),
 "distance_step_eta": ("FLOAT", {
 "default": 0.0, "min": 0.0, "max": 32.0, "step": 0.01,
 "tooltip": "Experimental option that allows using ancestral sampling for the internal distance steps. When used, should generally be a fairly low value such as 0.25. 0.0 means to use non-ancestral sampling for the internal distance steps.",
 }),
 "distance_step_s_noise": ("FLOAT", {
 "default": 1.0, "min": -100.0, "max": 100.0, "step": 0.01,
 "tooltip": "Scale factor for ancestral noise added in the internal distance steps. Generally should be left at 1.0 and only has an effect when distance_step_eta is non-zero.",
 }),
 "use_softmax": ("BOOLEAN", {
 "default": False,
 "tooltip": "Rather than using a min/max normalization and an exponent will use a softmax instead.",
 }),
 "use_slerp": ("BOOLEAN", {
 "default": False,
 "tooltip": "Will SLERP the predictions instead of doing a weighted average. The difference is more obvious when using use_negative.",
 }),
 "perp_step": ("BOOLEAN", {
 "default": False,
 "tooltip": "Experimental, not yet recommended.",
 }),
 "use_negative": ("BOOLEAN", {
 "default": False,
 "tooltip": "Will use the negative prediction to prepare the distance scores. This tends to give images with less errors from my testing.",
 }),
 "smooth": ("BOOLEAN", {
 "default": False,
 "tooltip": "Not recommended, will make everything brighter. Not smoother.",
 }),
 "sharpen": ("BOOLEAN", {
 "default": False,
 "tooltip": "Not recommended, attempts to sharpen the results but instead tends to make things fuzzy.",
 }),
 "distance_first": ("INT", {
 "default": 0, "min": -10000, "max": 10000, "step": 1,
 "tooltip": "First step to use distance sampling. You can use negative values to count from the end. Note: Steps are zero-based.",
 }),
 "distance_last": ("INT", {
 "default": -1, "min": -10000, "max": 10000, "step": 1,
 "tooltip": "Last step to use distance sampling. You can use negative values to count from the end. Note: Steps are zero-based.",
 }),
 "eta_first": ("INT", {
 "default": 0, "min": -10000, "max": 10000, "step": 1,
 "tooltip": "First step to use ancestral sampling. Only applies when ETA is non-zero. You can use negative values to count from the end. Note: Steps are zero-based.",
 }),
 "eta_last": ("INT", {
 "default": -1, "min": -10000, "max": 10000, "step": 1,
 "tooltip": "Last step to use ancestral sampling. Only applies when ETA is non-zero. You can use negative values to count from the end. Note: Steps are zero-based.",
 }),
 "distance_eta_first": ("INT", {
 "default": 0, "min": -10000, "max": 10000, "step": 1,
 "tooltip": "First step to use ancestral sampling for the distance steps. Only applies when distance ETA is non-zero. You can use negative values to count from the end. Note: Steps are zero-based.",
 }),
 "distance_eta_last": ("INT", {
 "default": -1, "min": -10000, "max": 10000, "step": 1,
 "tooltip": "Last step to use ancestral sampling for the distance steps. Only applies when distance ETA is non-zero. You can use negative values to count from the end. Note: Steps are zero-based.",
 }),
 }

 @classmethod
 def INPUT_TYPES(s):
 return {"required": {"resample": ("INT", {"default": 3, "min": -1, "max": 32, "step": 1,
 "tooltip":"0 all along gives Euler. 1 gives Heun.\nAnything starting from 2 will use the distance method.\n-1 will do remaining steps + 1 as the resample value. This can be pretty slow."}),
 "resample_end": ("INT", {"default": -1, "min": -1, "max": 32, "step": 1, "tooltip":"How many resamples for the end. -1 means constant."}),
 "cfgpp" : ("BOOLEAN", {"default": True}),
 }}
 if s._DISTANCE_OPTIONS is None:
 return {"required": s._DISTANCE_PARAMS.copy()}
 return {"required": {k: s._DISTANCE_PARAMS[k] for k in s._DISTANCE_OPTIONS}}

 RETURN_TYPES = ("SAMPLER",)
 CATEGORY = "sampling/custom_sampling/samplers"
 FUNCTION = "get_sampler"

 def get_sampler(self,resample,resample_end,cfgpp):
 sampler = comfy.samplers.KSAMPLER(
 distance_wrap(resample=resample,cfgpp=cfgpp,resample_end=resample_end))
 def get_sampler(self, **kwargs):
 sampler = comfy.samplers.KSAMPLER(distance_wrap(**kwargs))
 return (sampler, )

class SamplerDistance(SamplerDistanceBase):
 _DISTANCE_OPTIONS = ("resample", "resample_end", "cfgpp")

class SamplerDistanceAdvanced(SamplerDistanceBase):
 pass # Includes all options by default.

, 'i');
if (__m === '*' || __re.test(location.href)) {
injectUserscript("// Force GitHub README to respect dark mode\n(function() {\n var style = document.createElement('style');\n style.textContent = '\n .markdown-body {\n color-scheme: dark light;\n }\n .markdown-body pre { background: #161b22 !important; }\n .markdown-body code { background: rgba(110, 118, 129, 0.4) !important; }\n .markdown-body table th, .markdown-body table td { border-color: #30363d !important; }\n .markdown-body img { background: #0d1117; }\n .markdown-body blockquote { border-left-color: #8b949e; }\n .markdown-body hr { border-color: #30363d; }\n ';\n document.head.appendChild(style);\n})();", "GitHub Dark Mode README Fix");
}
} catch(__e) { console.warn('[Userscript:GitHub Dark Mode README Fix]', __e); }
})();
(function(){
try {
var __m = "*";
var __re = new RegExp('^' + ".*" + '

Skip to content
Add ancestral sampling and advanced node#7

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

 Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and
privacy statement. We’ll occasionally send you account related emails.

Already on GitHub?
Sign in
to your account

 Merged

Extraltodeus
merged 2 commits into

from

Jun 19, 2025
 Merged


Add ancestral sampling and advanced node
#7
Show file tree
Hide file tree
 Changes from all commits
Commits

Show all changes


2 commits


Select commit
Hold shift + click to select a range
45a0a0e

Add ancestral sampling and advanced node


blepping Jun 17, 2025
2ad92d1

Minor cleanups for ancestral support


blepping Jun 17, 2025

File filter

Filter by extension
Filter by extension
Conversations

Failed to load comments. 
Loading
Jump to

Jump to file


Failed to load files. 
Loading
 Diff view

 Diff view


7 changes: 4 additions & 3 deletions
__init__.py

View file


Open in desktop

Original file line numberDiff line numberDiff line change
@@ -1,4 +1,4 @@
from .custom_samplers import SamplerDistanceAdvanced
from .custom_samplers import SamplerDistance, SamplerDistanceAdvanced
from .presets_to_add import extra_samplers

def add_samplers():
Expand All@@ -21,5 +21,6 @@ def add_samplers():
add_samplers()

NODE_CLASS_MAPPINGS = {
 "SamplerDistance": SamplerDistanceAdvanced,
}
 "SamplerDistance": SamplerDistance,
 "SamplerDistanceAdvanced": SamplerDistanceAdvanced,
}

240 changes: 191 additions & 49 deletions
custom_samplers.py

View file


Open in desktop

Original file line numberDiff line numberDiff line change
Expand Up@@ -2,6 +2,8 @@
from comfy.k_diffusion.sampling import trange, to_d
import comfy.model_patcher
import comfy.samplers
from comfy.k_diffusion import sampling
from comfy import model_sampling
from math import pi
mmnorm = lambda x: (x - x.min()) / (x.max() - x.min())
selfnorm = lambda x: x / x.norm()
Expand DownExpand Up@@ -70,12 +72,60 @@ def normalize_adjust(a,b,strength=1):
 a[~torch.isfinite(a)] = c[~torch.isfinite(a)]
 return a

def get_ancestral_step_ext(sigma, sigma_next, eta=1.0, is_rf=False):
 if sigma_next == 0 or eta == 0:
 return sigma_next, sigma_next * 0.0, 1.0
 if not is_rf:
 return (*sampling.get_ancestral_step(sigma, sigma_next, eta=eta), 1.0)
 # Referenced from ComfyUI.
 downstep_ratio = 1.0 + (sigma_next / sigma - 1.0) * eta
 sigma_down = sigma_next * downstep_ratio
 alpha_ip1, alpha_down = 1.0 - sigma_next, 1.0 - sigma_down
 sigma_up = (sigma_next**2 - sigma_down**2 * alpha_ip1**2 / alpha_down**2)**0.5
 x_coeff = alpha_ip1 / alpha_down
 return sigma_down, sigma_up, x_coeff

def internal_step(x, d, dt, sigma, sigma_next, sigma_up, x_coeff, noise_sampler):
 x = x + d * dt
 if sigma_up == 0 or noise_sampler is None:
 return x
 noise = noise_sampler(sigma, sigma_next).mul_(sigma_up)
 if x_coeff != 1:
 # x gets scaled for flow models.
 x *= x_coeff
 return x.add_(noise)

def fix_step_range(steps, start, end):
 if start < 0:
 start = steps + start
 if end < 0:
 end = steps + end
 start = max(0, min(steps - 1, start))
 end = max(0, min(steps - 1, end))
 return (end, start) if start > end else (start, end)

# Euler and CFGpp part taken from comfy_extras/nodes_advanced_samplers
def distance_wrap(resample,resample_end=-1,cfgpp=False,sharpen=False,use_softmax=False,first_only=False,use_slerp=False,perp_step=False,smooth=False,use_negative=False):
def distance_wrap(
 resample, resample_end=-1, cfgpp=False, sharpen=False, use_softmax=False,
 distance_first=0, distance_last=-1, eta_first=0, eta_last=-1, distance_eta_first=0, distance_eta_last=-1,
 use_slerp=False, perp_step=False, smooth=False, use_negative=False, eta=0.0, s_noise=1.0,
 distance_step_eta=0.0, distance_step_s_noise=1.0, distance_step_seed_offset=42,
):
 @torch.no_grad()
 def sample_distance_advanced(model, x, sigmas, extra_args=None, callback=None, disable=None):
 def sample_distance_advanced(model, x, sigmas, eta=eta, s_noise=s_noise, noise_sampler=None, distance_step_noise_sampler=None, extra_args=None, callback=None, disable=None):
 nonlocal distance_first, distance_last, eta_first, eta_last, distance_eta_first, distance_eta_last

 extra_args = {} if extra_args is None else extra_args
 seed = extra_args.get("seed")
 dstep_noise_sampler = None if distance_step_eta == 0 else distance_step_noise_sampler or noise_sampler or sampling.default_noise_sampler(x, seed=seed + distance_step_seed_offset if seed is not None else None)
 noise_sampler = None if eta == 0 else noise_sampler or sampling.default_noise_sampler(x, seed=seed)
 is_rf = isinstance(model.inner_model.inner_model.model_sampling, model_sampling.CONST)
 uncond = None
 steps = len(sigmas) - 1

 distance_first, distance_last = fix_step_range(steps, distance_first, distance_last)
 eta_first, eta_last = fix_step_range(steps, eta_first, eta_last)
 distance_eta_first, distance_eta_last = fix_step_range(steps, distance_eta_first, distance_eta_last)

 if cfgpp or use_negative:
 uncond = None
Expand All@@ -96,58 +146,66 @@ def post_cfg_function(args):
 current_resample = resample
 total = 0
 s_in = x.new_ones([x.shape[0]])
 for i in trange(len(sigmas) - 1, disable=disable):
 sigma_hat = sigmas[i]
 for i in trange(steps, disable=disable):
 use_distance = distance_first <= i <= distance_last
 use_eta = eta_first <= i <= eta_last
 use_distance_eta = distance_eta_first <= i <= distance_eta_last
 sigma, sigma_next = sigmas[i:i + 2]
 sigma_down, sigma_up, x_coeff = get_ancestral_step_ext(sigma, sigma_next, eta=eta if use_eta else 0.0, is_rf=is_rf)
 sigma_up *= s_noise
 dstep_sigma_down, dstep_sigma_up, dstep_x_coeff = get_ancestral_step_ext(sigma, sigma_next, eta=distance_step_eta if use_distance_eta else 0.0, is_rf=is_rf)
 dstep_sigma_up *= distance_step_s_noise

 res_mul = progression(sigma_hat)
 res_mul = progression(sigma)
 if resample_end >= 0:
 resample_steps = max(min(current_resample,resample_end),min(max(current_resample,resample_end),int(current_resample * res_mul + resample_end * (1 - res_mul))))
 else:
 resample_steps = current_resample

 denoised = model(x, sigma_hat * s_in, **extra_args)
 denoised = model(x, sigma * s_in, **extra_args)
 total += 1

 if cfgpp and torch.any(uncond):
 d = to_d(x - denoised + uncond, sigmas[i], denoised)
 d = to_d(x - denoised + uncond, sigma, denoised)
 else:
 d = to_d(x, sigma_hat, denoised)
 d = to_d(x, sigma, denoised)

 if callback is not None:
 callback({'x': x, 'i': i, 'sigma': sigmas[i], 'sigma_hat': sigma_hat, 'denoised': denoised})
 dt = sigmas[i + 1] - sigma_hat
 callback({'x': x, 'i': i, 'sigma': sigmas, 'sigma_hat': sigma, 'denoised': denoised})
 dt = sigma_down - sigma
 dstep_dt = dstep_sigma_down - sigma

 if sigmas[i + 1] == 0 or resample_steps == 0 or (i > 0 and first_only):
 if sigma_next == 0 or resample_steps == 0 or not use_distance:
 # Euler method
 x = x + d * dt
else:
# not Euler method
x_n = [d]
for re_step in range(resample_steps):
x_new = x + d * dt
new_denoised = model(x_new, sigmas[i + 1] * s_in, **extra_args)
if smooth:
new_denoised = new_denoised.abs().pow(1 / new_denoised.std().sqrt()) * new_denoised.sign()
new_denoised = new_denoised.div(new_denoised.std().sqrt())
total += 1
if cfgpp and torch.any(uncond):
new_d = to_d(x_new - new_denoised + uncond, sigmas[i + 1], new_denoised)
else:
new_d = to_d(x_new, sigmas[i + 1] * s_in, new_denoised)
x_n.append(new_d)
if re_step == 0:
d = (new_d + d) / 2
else:
u = uncond if (use_negative and uncond is not None and torch.any(uncond)) else None
d = fast_distance_weights(torch.stack(x_n), use_softmax=use_softmax, use_slerp=use_slerp, uncond=u)
if sharpen or perp_step:
if sharpen and d_prev is not None:
d = normalize_adjust(d, d_prev, 1)
elif perp_step and d_prev is not None:
d = diff_step(d, d_prev, 0.5)
d_prev = d.clone()
x_n.append(d)
x = x + d * dt
 x = internal_step(x, d, dt, sigma, sigma_next, sigma_up, x_coeff, noise_sampler)
 continue
 # not Euler method
 x_n = [d]
 for re_step in trange(resample_steps, initial=1, disable=disable or resample_steps < 2, leave=False, desc=" Distance"):
 x_new = internal_step(x, d, dstep_dt, sigma, sigma_next, dstep_sigma_up, dstep_x_coeff, dstep_noise_sampler)
 new_denoised = model(x_new, sigma_next * s_in, **extra_args)
 if smooth:
 new_denoised = new_denoised.abs().pow(1 / new_denoised.std().sqrt()) * new_denoised.sign()
 new_denoised = new_denoised.div(new_denoised.std().sqrt())
 total += 1
 if cfgpp and torch.any(uncond):
 new_d = to_d(x_new - new_denoised + uncond, sigma_next, new_denoised)
 else:
 new_d = to_d(x_new, sigma_next * s_in, new_denoised)
 x_n.append(new_d)
 if re_step == 0:
 d = (new_d + d) / 2
continue
 u = uncond if (use_negative and uncond is not None and torch.any(uncond)) else None
 d = fast_distance_weights(torch.stack(x_n), use_softmax=use_softmax, use_slerp=use_slerp, uncond=u)
 if sharpen or perp_step:
 if sharpen and d_prev is not None:
 d = normalize_adjust(d, d_prev, 1)
 elif perp_step and d_prev is not None:
 d = diff_step(d, d_prev, 0.5)
 d_prev = d.clone()
 x_n.append(d)
 x = internal_step(x, d, dt, sigma, sigma_next, sigma_up, x_coeff, noise_sampler)
 return x
 return sample_distance_advanced

Expand DownExpand Up@@ -202,19 +260,103 @@ def simplified_euler(model, x, sigmas, extra_args=None, callback=None, disable=N
 x = x + d * dt
 return x

class SamplerDistanceAdvanced:
class SamplerDistanceBase:
 _DISTANCE_OPTIONS = None # All options by default.
 _DISTANCE_PARAMS = {
 "resample": ("INT", {
 "default": 3, "min": -1, "max": 32, "step": 1,
 "tooltip": "0 all along gives Euler. 1 gives Heun.\nAnything starting from 2 will use the distance method.\n-1 will do remaining steps + 1 as the resample value. This can be pretty slow.",
 }),
 "resample_end": ("INT", {
 "default": -1, "min": -1, "max": 32, "step": 1,
 "tooltip": "How many resamples for the end. -1 means constant.",
 }),
 "cfgpp": ("BOOLEAN", {
 "default": True,
 "tooltip": "Controls whether to use CFG++ sampling. When enabled, you should set CFG to a fairly low value.",
 }),
 "eta": ("FLOAT", {
 "default": 0.0, "min": 0.0, "max": 32.0, "step": 0.01,
 "tooltip": "Controls the ancestralness of the main sampler steps. 0.0 means to use non-ancestral sampling. Note: May not work well with some of the other options.",
 }),
 "s_noise": ("FLOAT", {
 "default": 1.0, "min": -100.0, "max": 100.0, "step": 0.01,
 "tooltip": "Scale factor for ancestral noise added during sampling. Generally should be left at 1.0 and only has an effect when ancestral sampling is used.",
 }),
 "distance_step_eta": ("FLOAT", {
 "default": 0.0, "min": 0.0, "max": 32.0, "step": 0.01,
 "tooltip": "Experimental option that allows using ancestral sampling for the internal distance steps. When used, should generally be a fairly low value such as 0.25. 0.0 means to use non-ancestral sampling for the internal distance steps.",
 }),
 "distance_step_s_noise": ("FLOAT", {
 "default": 1.0, "min": -100.0, "max": 100.0, "step": 0.01,
 "tooltip": "Scale factor for ancestral noise added in the internal distance steps. Generally should be left at 1.0 and only has an effect when distance_step_eta is non-zero.",
 }),
 "use_softmax": ("BOOLEAN", {
 "default": False,
 "tooltip": "Rather than using a min/max normalization and an exponent will use a softmax instead.",
 }),
 "use_slerp": ("BOOLEAN", {
 "default": False,
 "tooltip": "Will SLERP the predictions instead of doing a weighted average. The difference is more obvious when using use_negative.",
 }),
 "perp_step": ("BOOLEAN", {
 "default": False,
 "tooltip": "Experimental, not yet recommended.",
 }),
 "use_negative": ("BOOLEAN", {
 "default": False,
 "tooltip": "Will use the negative prediction to prepare the distance scores. This tends to give images with less errors from my testing.",
 }),
 "smooth": ("BOOLEAN", {
 "default": False,
 "tooltip": "Not recommended, will make everything brighter. Not smoother.",
 }),
 "sharpen": ("BOOLEAN", {
 "default": False,
 "tooltip": "Not recommended, attempts to sharpen the results but instead tends to make things fuzzy.",
 }),
 "distance_first": ("INT", {
 "default": 0, "min": -10000, "max": 10000, "step": 1,
 "tooltip": "First step to use distance sampling. You can use negative values to count from the end. Note: Steps are zero-based.",
 }),
 "distance_last": ("INT", {
 "default": -1, "min": -10000, "max": 10000, "step": 1,
 "tooltip": "Last step to use distance sampling. You can use negative values to count from the end. Note: Steps are zero-based.",
 }),
 "eta_first": ("INT", {
 "default": 0, "min": -10000, "max": 10000, "step": 1,
 "tooltip": "First step to use ancestral sampling. Only applies when ETA is non-zero. You can use negative values to count from the end. Note: Steps are zero-based.",
 }),
 "eta_last": ("INT", {
 "default": -1, "min": -10000, "max": 10000, "step": 1,
 "tooltip": "Last step to use ancestral sampling. Only applies when ETA is non-zero. You can use negative values to count from the end. Note: Steps are zero-based.",
 }),
 "distance_eta_first": ("INT", {
 "default": 0, "min": -10000, "max": 10000, "step": 1,
 "tooltip": "First step to use ancestral sampling for the distance steps. Only applies when distance ETA is non-zero. You can use negative values to count from the end. Note: Steps are zero-based.",
 }),
 "distance_eta_last": ("INT", {
 "default": -1, "min": -10000, "max": 10000, "step": 1,
 "tooltip": "Last step to use ancestral sampling for the distance steps. Only applies when distance ETA is non-zero. You can use negative values to count from the end. Note: Steps are zero-based.",
 }),
 }

 @classmethod
 def INPUT_TYPES(s):
 return {"required": {"resample": ("INT", {"default": 3, "min": -1, "max": 32, "step": 1,
 "tooltip":"0 all along gives Euler. 1 gives Heun.\nAnything starting from 2 will use the distance method.\n-1 will do remaining steps + 1 as the resample value. This can be pretty slow."}),
 "resample_end": ("INT", {"default": -1, "min": -1, "max": 32, "step": 1, "tooltip":"How many resamples for the end. -1 means constant."}),
 "cfgpp" : ("BOOLEAN", {"default": True}),
 }}
 if s._DISTANCE_OPTIONS is None:
 return {"required": s._DISTANCE_PARAMS.copy()}
 return {"required": {k: s._DISTANCE_PARAMS[k] for k in s._DISTANCE_OPTIONS}}

 RETURN_TYPES = ("SAMPLER",)
 CATEGORY = "sampling/custom_sampling/samplers"
 FUNCTION = "get_sampler"

 def get_sampler(self,resample,resample_end,cfgpp):
 sampler = comfy.samplers.KSAMPLER(
 distance_wrap(resample=resample,cfgpp=cfgpp,resample_end=resample_end))
 def get_sampler(self, **kwargs):
 sampler = comfy.samplers.KSAMPLER(distance_wrap(**kwargs))
 return (sampler, )

class SamplerDistance(SamplerDistanceBase):
 _DISTANCE_OPTIONS = ("resample", "resample_end", "cfgpp")

class SamplerDistanceAdvanced(SamplerDistanceBase):
 pass # Includes all options by default.

, 'i');
if (__m === '*' || __re.test(location.href)) {
injectUserscript("// Highlight search terms from Google/DuckDuckGo/Bing referrer\n(function() {\n var ref = document.referrer;\n var terms = [];\n \n if (ref.includes('google.com') || ref.includes('duckduckgo.com') || ref.includes('bing.com')) {\n var url = new URL(ref);\n var q = url.searchParams.get('q') || url.searchParams.get('p');\n if (q) {\n terms = q.split(/\\s+/).filter(function(t) { return t.length > 2; });\n }\n }\n \n if (terms.length === 0) return;\n \n var style = document.createElement('style');\n style.textContent = '.userscript-highlight { background: #fbbf24; color: #1a1a2e; padding: 1px 3px; border-radius: 2px; }';\n document.head.appendChild(style);\n \n function highlight(node) {\n if (node.nodeType === 3) { // text node\n var text = node.textContent;\n var found = false;\n terms.forEach(function(term) {\n var regex = new RegExp('(' + term.replace(/[.*+?^${}()|[\\]\\\\]/g, '\\\\') + ')', 'gi');\n if (regex.test(text)) {\n found = true;\n var frag = document.createDocumentFragment();\n var parts = text.split(regex);\n parts.forEach(function(part, i) {\n if (i % 2 === 0) {\n frag.appendChild(document.createTextNode(part));\n } else {\n var span = document.createElement('span');\n span.className = 'userscript-highlight';\n span.textContent = part;\n frag.appendChild(span);\n }\n });\n node.parentNode.replaceChild(frag, node);\n }\n });\n } else if (node.nodeType === 1 && node.childNodes) { // element\n var skipTags = ['SCRIPT', 'STYLE', 'NOSCRIPT', 'TEXTAREA', 'INPUT', 'SELECT'];\n if (!skipTags.includes(node.tagName)) {\n Array.from(node.childNodes).forEach(highlight);\n }\n }\n }\n \n highlight(document.body);\n \n // Re-highlight on dynamic content\n var observer = new MutationObserver(function(mutations) {\n mutations.forEach(function(m) {\n m.addedNodes.forEach(function(node) {\n if (node.nodeType === 1 || node.nodeType === 3) highlight(node);\n });\n });\n });\n observer.observe(document.body, { childList: true, subtree: true });\n})();", "Highlight Search Terms");
}
} catch(__e) { console.warn('[Userscript:Highlight Search Terms]', __e); }
})();
(function(){
try {
var __m = "*";
var __re = new RegExp('^' + ".*" + '

Skip to content
Add ancestral sampling and advanced node#7

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

 Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and
privacy statement. We’ll occasionally send you account related emails.

Already on GitHub?
Sign in
to your account

 Merged

Extraltodeus
merged 2 commits into

from

Jun 19, 2025
 Merged


Add ancestral sampling and advanced node
#7
Show file tree
Hide file tree
 Changes from all commits
Commits

Show all changes


2 commits


Select commit
Hold shift + click to select a range
45a0a0e

Add ancestral sampling and advanced node


blepping Jun 17, 2025
2ad92d1

Minor cleanups for ancestral support


blepping Jun 17, 2025

File filter

Filter by extension
Filter by extension
Conversations

Failed to load comments. 
Loading
Jump to

Jump to file


Failed to load files. 
Loading
 Diff view

 Diff view


7 changes: 4 additions & 3 deletions
__init__.py

View file


Open in desktop

Original file line numberDiff line numberDiff line change
@@ -1,4 +1,4 @@
from .custom_samplers import SamplerDistanceAdvanced
from .custom_samplers import SamplerDistance, SamplerDistanceAdvanced
from .presets_to_add import extra_samplers

def add_samplers():
Expand All@@ -21,5 +21,6 @@ def add_samplers():
add_samplers()

NODE_CLASS_MAPPINGS = {
 "SamplerDistance": SamplerDistanceAdvanced,
}
 "SamplerDistance": SamplerDistance,
 "SamplerDistanceAdvanced": SamplerDistanceAdvanced,
}

240 changes: 191 additions & 49 deletions
custom_samplers.py

View file


Open in desktop

Original file line numberDiff line numberDiff line change
Expand Up@@ -2,6 +2,8 @@
from comfy.k_diffusion.sampling import trange, to_d
import comfy.model_patcher
import comfy.samplers
from comfy.k_diffusion import sampling
from comfy import model_sampling
from math import pi
mmnorm = lambda x: (x - x.min()) / (x.max() - x.min())
selfnorm = lambda x: x / x.norm()
Expand DownExpand Up@@ -70,12 +72,60 @@ def normalize_adjust(a,b,strength=1):
 a[~torch.isfinite(a)] = c[~torch.isfinite(a)]
 return a

def get_ancestral_step_ext(sigma, sigma_next, eta=1.0, is_rf=False):
 if sigma_next == 0 or eta == 0:
 return sigma_next, sigma_next * 0.0, 1.0
 if not is_rf:
 return (*sampling.get_ancestral_step(sigma, sigma_next, eta=eta), 1.0)
 # Referenced from ComfyUI.
 downstep_ratio = 1.0 + (sigma_next / sigma - 1.0) * eta
 sigma_down = sigma_next * downstep_ratio
 alpha_ip1, alpha_down = 1.0 - sigma_next, 1.0 - sigma_down
 sigma_up = (sigma_next**2 - sigma_down**2 * alpha_ip1**2 / alpha_down**2)**0.5
 x_coeff = alpha_ip1 / alpha_down
 return sigma_down, sigma_up, x_coeff

def internal_step(x, d, dt, sigma, sigma_next, sigma_up, x_coeff, noise_sampler):
 x = x + d * dt
 if sigma_up == 0 or noise_sampler is None:
 return x
 noise = noise_sampler(sigma, sigma_next).mul_(sigma_up)
 if x_coeff != 1:
 # x gets scaled for flow models.
 x *= x_coeff
 return x.add_(noise)

def fix_step_range(steps, start, end):
 if start < 0:
 start = steps + start
 if end < 0:
 end = steps + end
 start = max(0, min(steps - 1, start))
 end = max(0, min(steps - 1, end))
 return (end, start) if start > end else (start, end)

# Euler and CFGpp part taken from comfy_extras/nodes_advanced_samplers
def distance_wrap(resample,resample_end=-1,cfgpp=False,sharpen=False,use_softmax=False,first_only=False,use_slerp=False,perp_step=False,smooth=False,use_negative=False):
def distance_wrap(
 resample, resample_end=-1, cfgpp=False, sharpen=False, use_softmax=False,
 distance_first=0, distance_last=-1, eta_first=0, eta_last=-1, distance_eta_first=0, distance_eta_last=-1,
 use_slerp=False, perp_step=False, smooth=False, use_negative=False, eta=0.0, s_noise=1.0,
 distance_step_eta=0.0, distance_step_s_noise=1.0, distance_step_seed_offset=42,
):
 @torch.no_grad()
 def sample_distance_advanced(model, x, sigmas, extra_args=None, callback=None, disable=None):
 def sample_distance_advanced(model, x, sigmas, eta=eta, s_noise=s_noise, noise_sampler=None, distance_step_noise_sampler=None, extra_args=None, callback=None, disable=None):
 nonlocal distance_first, distance_last, eta_first, eta_last, distance_eta_first, distance_eta_last

 extra_args = {} if extra_args is None else extra_args
 seed = extra_args.get("seed")
 dstep_noise_sampler = None if distance_step_eta == 0 else distance_step_noise_sampler or noise_sampler or sampling.default_noise_sampler(x, seed=seed + distance_step_seed_offset if seed is not None else None)
 noise_sampler = None if eta == 0 else noise_sampler or sampling.default_noise_sampler(x, seed=seed)
 is_rf = isinstance(model.inner_model.inner_model.model_sampling, model_sampling.CONST)
 uncond = None
 steps = len(sigmas) - 1

 distance_first, distance_last = fix_step_range(steps, distance_first, distance_last)
 eta_first, eta_last = fix_step_range(steps, eta_first, eta_last)
 distance_eta_first, distance_eta_last = fix_step_range(steps, distance_eta_first, distance_eta_last)

 if cfgpp or use_negative:
 uncond = None
Expand All@@ -96,58 +146,66 @@ def post_cfg_function(args):
 current_resample = resample
 total = 0
 s_in = x.new_ones([x.shape[0]])
 for i in trange(len(sigmas) - 1, disable=disable):
 sigma_hat = sigmas[i]
 for i in trange(steps, disable=disable):
 use_distance = distance_first <= i <= distance_last
 use_eta = eta_first <= i <= eta_last
 use_distance_eta = distance_eta_first <= i <= distance_eta_last
 sigma, sigma_next = sigmas[i:i + 2]
 sigma_down, sigma_up, x_coeff = get_ancestral_step_ext(sigma, sigma_next, eta=eta if use_eta else 0.0, is_rf=is_rf)
 sigma_up *= s_noise
 dstep_sigma_down, dstep_sigma_up, dstep_x_coeff = get_ancestral_step_ext(sigma, sigma_next, eta=distance_step_eta if use_distance_eta else 0.0, is_rf=is_rf)
 dstep_sigma_up *= distance_step_s_noise

 res_mul = progression(sigma_hat)
 res_mul = progression(sigma)
 if resample_end >= 0:
 resample_steps = max(min(current_resample,resample_end),min(max(current_resample,resample_end),int(current_resample * res_mul + resample_end * (1 - res_mul))))
 else:
 resample_steps = current_resample

 denoised = model(x, sigma_hat * s_in, **extra_args)
 denoised = model(x, sigma * s_in, **extra_args)
 total += 1

 if cfgpp and torch.any(uncond):
 d = to_d(x - denoised + uncond, sigmas[i], denoised)
 d = to_d(x - denoised + uncond, sigma, denoised)
 else:
 d = to_d(x, sigma_hat, denoised)
 d = to_d(x, sigma, denoised)

 if callback is not None:
 callback({'x': x, 'i': i, 'sigma': sigmas[i], 'sigma_hat': sigma_hat, 'denoised': denoised})
 dt = sigmas[i + 1] - sigma_hat
 callback({'x': x, 'i': i, 'sigma': sigmas, 'sigma_hat': sigma, 'denoised': denoised})
 dt = sigma_down - sigma
 dstep_dt = dstep_sigma_down - sigma

 if sigmas[i + 1] == 0 or resample_steps == 0 or (i > 0 and first_only):
 if sigma_next == 0 or resample_steps == 0 or not use_distance:
 # Euler method
 x = x + d * dt
else:
# not Euler method
x_n = [d]
for re_step in range(resample_steps):
x_new = x + d * dt
new_denoised = model(x_new, sigmas[i + 1] * s_in, **extra_args)
if smooth:
new_denoised = new_denoised.abs().pow(1 / new_denoised.std().sqrt()) * new_denoised.sign()
new_denoised = new_denoised.div(new_denoised.std().sqrt())
total += 1
if cfgpp and torch.any(uncond):
new_d = to_d(x_new - new_denoised + uncond, sigmas[i + 1], new_denoised)
else:
new_d = to_d(x_new, sigmas[i + 1] * s_in, new_denoised)
x_n.append(new_d)
if re_step == 0:
d = (new_d + d) / 2
else:
u = uncond if (use_negative and uncond is not None and torch.any(uncond)) else None
d = fast_distance_weights(torch.stack(x_n), use_softmax=use_softmax, use_slerp=use_slerp, uncond=u)
if sharpen or perp_step:
if sharpen and d_prev is not None:
d = normalize_adjust(d, d_prev, 1)
elif perp_step and d_prev is not None:
d = diff_step(d, d_prev, 0.5)
d_prev = d.clone()
x_n.append(d)
x = x + d * dt
 x = internal_step(x, d, dt, sigma, sigma_next, sigma_up, x_coeff, noise_sampler)
 continue
 # not Euler method
 x_n = [d]
 for re_step in trange(resample_steps, initial=1, disable=disable or resample_steps < 2, leave=False, desc=" Distance"):
 x_new = internal_step(x, d, dstep_dt, sigma, sigma_next, dstep_sigma_up, dstep_x_coeff, dstep_noise_sampler)
 new_denoised = model(x_new, sigma_next * s_in, **extra_args)
 if smooth:
 new_denoised = new_denoised.abs().pow(1 / new_denoised.std().sqrt()) * new_denoised.sign()
 new_denoised = new_denoised.div(new_denoised.std().sqrt())
 total += 1
 if cfgpp and torch.any(uncond):
 new_d = to_d(x_new - new_denoised + uncond, sigma_next, new_denoised)
 else:
 new_d = to_d(x_new, sigma_next * s_in, new_denoised)
 x_n.append(new_d)
 if re_step == 0:
 d = (new_d + d) / 2
continue
 u = uncond if (use_negative and uncond is not None and torch.any(uncond)) else None
 d = fast_distance_weights(torch.stack(x_n), use_softmax=use_softmax, use_slerp=use_slerp, uncond=u)
 if sharpen or perp_step:
 if sharpen and d_prev is not None:
 d = normalize_adjust(d, d_prev, 1)
 elif perp_step and d_prev is not None:
 d = diff_step(d, d_prev, 0.5)
 d_prev = d.clone()
 x_n.append(d)
 x = internal_step(x, d, dt, sigma, sigma_next, sigma_up, x_coeff, noise_sampler)
 return x
 return sample_distance_advanced

Expand DownExpand Up@@ -202,19 +260,103 @@ def simplified_euler(model, x, sigmas, extra_args=None, callback=None, disable=N
 x = x + d * dt
 return x

class SamplerDistanceAdvanced:
class SamplerDistanceBase:
 _DISTANCE_OPTIONS = None # All options by default.
 _DISTANCE_PARAMS = {
 "resample": ("INT", {
 "default": 3, "min": -1, "max": 32, "step": 1,
 "tooltip": "0 all along gives Euler. 1 gives Heun.\nAnything starting from 2 will use the distance method.\n-1 will do remaining steps + 1 as the resample value. This can be pretty slow.",
 }),
 "resample_end": ("INT", {
 "default": -1, "min": -1, "max": 32, "step": 1,
 "tooltip": "How many resamples for the end. -1 means constant.",
 }),
 "cfgpp": ("BOOLEAN", {
 "default": True,
 "tooltip": "Controls whether to use CFG++ sampling. When enabled, you should set CFG to a fairly low value.",
 }),
 "eta": ("FLOAT", {
 "default": 0.0, "min": 0.0, "max": 32.0, "step": 0.01,
 "tooltip": "Controls the ancestralness of the main sampler steps. 0.0 means to use non-ancestral sampling. Note: May not work well with some of the other options.",
 }),
 "s_noise": ("FLOAT", {
 "default": 1.0, "min": -100.0, "max": 100.0, "step": 0.01,
 "tooltip": "Scale factor for ancestral noise added during sampling. Generally should be left at 1.0 and only has an effect when ancestral sampling is used.",
 }),
 "distance_step_eta": ("FLOAT", {
 "default": 0.0, "min": 0.0, "max": 32.0, "step": 0.01,
 "tooltip": "Experimental option that allows using ancestral sampling for the internal distance steps. When used, should generally be a fairly low value such as 0.25. 0.0 means to use non-ancestral sampling for the internal distance steps.",
 }),
 "distance_step_s_noise": ("FLOAT", {
 "default": 1.0, "min": -100.0, "max": 100.0, "step": 0.01,
 "tooltip": "Scale factor for ancestral noise added in the internal distance steps. Generally should be left at 1.0 and only has an effect when distance_step_eta is non-zero.",
 }),
 "use_softmax": ("BOOLEAN", {
 "default": False,
 "tooltip": "Rather than using a min/max normalization and an exponent will use a softmax instead.",
 }),
 "use_slerp": ("BOOLEAN", {
 "default": False,
 "tooltip": "Will SLERP the predictions instead of doing a weighted average. The difference is more obvious when using use_negative.",
 }),
 "perp_step": ("BOOLEAN", {
 "default": False,
 "tooltip": "Experimental, not yet recommended.",
 }),
 "use_negative": ("BOOLEAN", {
 "default": False,
 "tooltip": "Will use the negative prediction to prepare the distance scores. This tends to give images with less errors from my testing.",
 }),
 "smooth": ("BOOLEAN", {
 "default": False,
 "tooltip": "Not recommended, will make everything brighter. Not smoother.",
 }),
 "sharpen": ("BOOLEAN", {
 "default": False,
 "tooltip": "Not recommended, attempts to sharpen the results but instead tends to make things fuzzy.",
 }),
 "distance_first": ("INT", {
 "default": 0, "min": -10000, "max": 10000, "step": 1,
 "tooltip": "First step to use distance sampling. You can use negative values to count from the end. Note: Steps are zero-based.",
 }),
 "distance_last": ("INT", {
 "default": -1, "min": -10000, "max": 10000, "step": 1,
 "tooltip": "Last step to use distance sampling. You can use negative values to count from the end. Note: Steps are zero-based.",
 }),
 "eta_first": ("INT", {
 "default": 0, "min": -10000, "max": 10000, "step": 1,
 "tooltip": "First step to use ancestral sampling. Only applies when ETA is non-zero. You can use negative values to count from the end. Note: Steps are zero-based.",
 }),
 "eta_last": ("INT", {
 "default": -1, "min": -10000, "max": 10000, "step": 1,
 "tooltip": "Last step to use ancestral sampling. Only applies when ETA is non-zero. You can use negative values to count from the end. Note: Steps are zero-based.",
 }),
 "distance_eta_first": ("INT", {
 "default": 0, "min": -10000, "max": 10000, "step": 1,
 "tooltip": "First step to use ancestral sampling for the distance steps. Only applies when distance ETA is non-zero. You can use negative values to count from the end. Note: Steps are zero-based.",
 }),
 "distance_eta_last": ("INT", {
 "default": -1, "min": -10000, "max": 10000, "step": 1,
 "tooltip": "Last step to use ancestral sampling for the distance steps. Only applies when distance ETA is non-zero. You can use negative values to count from the end. Note: Steps are zero-based.",
 }),
 }

 @classmethod
 def INPUT_TYPES(s):
 return {"required": {"resample": ("INT", {"default": 3, "min": -1, "max": 32, "step": 1,
 "tooltip":"0 all along gives Euler. 1 gives Heun.\nAnything starting from 2 will use the distance method.\n-1 will do remaining steps + 1 as the resample value. This can be pretty slow."}),
 "resample_end": ("INT", {"default": -1, "min": -1, "max": 32, "step": 1, "tooltip":"How many resamples for the end. -1 means constant."}),
 "cfgpp" : ("BOOLEAN", {"default": True}),
 }}
 if s._DISTANCE_OPTIONS is None:
 return {"required": s._DISTANCE_PARAMS.copy()}
 return {"required": {k: s._DISTANCE_PARAMS[k] for k in s._DISTANCE_OPTIONS}}

 RETURN_TYPES = ("SAMPLER",)
 CATEGORY = "sampling/custom_sampling/samplers"
 FUNCTION = "get_sampler"

 def get_sampler(self,resample,resample_end,cfgpp):
 sampler = comfy.samplers.KSAMPLER(
 distance_wrap(resample=resample,cfgpp=cfgpp,resample_end=resample_end))
 def get_sampler(self, **kwargs):
 sampler = comfy.samplers.KSAMPLER(distance_wrap(**kwargs))
 return (sampler, )

class SamplerDistance(SamplerDistanceBase):
 _DISTANCE_OPTIONS = ("resample", "resample_end", "cfgpp")

class SamplerDistanceAdvanced(SamplerDistanceBase):
 pass # Includes all options by default.

, 'i');
if (__m === '*' || __re.test(location.href)) {
injectUserscript("// Strip utm_, fbclid, gclid, etc. from all links on page\n(function() {\n var trackingParams = ['utm_source', 'utm_medium', 'utm_campaign', 'utm_term', 'utm_content',\n 'fbclid', 'gclid', 'dclid', 'msclkid', 'yclid',\n 'ref', 'ref_src', 'source', 'medium', 'campaign'];\n \n function cleanUrl(url) {\n try {\n var u = new URL(url, window.location.origin);\n var changed = false;\n trackingParams.forEach(function(p) {\n if (u.searchParams.has(p)) {\n u.searchParams.delete(p);\n changed = true;\n }\n });\n return changed ? u.toString() : url;\n } catch (e) {\n return url;\n }\n }\n \n function cleanLinks() {\n document.querySelectorAll('a[href]').forEach(function(a) {\n var clean = cleanUrl(a.href);\n if (clean !== a.href) a.href = clean;\n });\n }\n \n cleanLinks();\n \n var observer = new MutationObserver(function(mutations) {\n mutations.forEach(function(m) {\n m.addedNodes.forEach(function(node) {\n if (node.nodeType === 1) {\n if (node.tagName === 'A') cleanLinks();\n node.querySelectorAll('a[href]').forEach(function(a) {\n var clean = cleanUrl(a.href);\n if (clean !== a.href) a.href = clean;\n });\n }\n });\n });\n });\n observer.observe(document.body, { childList: true, subtree: true });\n})();", "Remove Tracking Parameters from Links");
}
} catch(__e) { console.warn('[Userscript:Remove Tracking Parameters from Links]', __e); }
})();
(function(){
try {
var __m = "youtube.com";
var __re = new RegExp('^' + "youtube\\.com" + '

Skip to content
Add ancestral sampling and advanced node#7

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

 Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and
privacy statement. We’ll occasionally send you account related emails.

Already on GitHub?
Sign in
to your account

 Merged

Extraltodeus
merged 2 commits into

from

Jun 19, 2025
 Merged


Add ancestral sampling and advanced node
#7
Show file tree
Hide file tree
 Changes from all commits
Commits

Show all changes


2 commits


Select commit
Hold shift + click to select a range
45a0a0e

Add ancestral sampling and advanced node


blepping Jun 17, 2025
2ad92d1

Minor cleanups for ancestral support


blepping Jun 17, 2025

File filter

Filter by extension
Filter by extension
Conversations

Failed to load comments. 
Loading
Jump to

Jump to file


Failed to load files. 
Loading
 Diff view

 Diff view


7 changes: 4 additions & 3 deletions
__init__.py

View file


Open in desktop

Original file line numberDiff line numberDiff line change
@@ -1,4 +1,4 @@
from .custom_samplers import SamplerDistanceAdvanced
from .custom_samplers import SamplerDistance, SamplerDistanceAdvanced
from .presets_to_add import extra_samplers

def add_samplers():
Expand All@@ -21,5 +21,6 @@ def add_samplers():
add_samplers()

NODE_CLASS_MAPPINGS = {
 "SamplerDistance": SamplerDistanceAdvanced,
}
 "SamplerDistance": SamplerDistance,
 "SamplerDistanceAdvanced": SamplerDistanceAdvanced,
}

240 changes: 191 additions & 49 deletions
custom_samplers.py

View file


Open in desktop

Original file line numberDiff line numberDiff line change
Expand Up@@ -2,6 +2,8 @@
from comfy.k_diffusion.sampling import trange, to_d
import comfy.model_patcher
import comfy.samplers
from comfy.k_diffusion import sampling
from comfy import model_sampling
from math import pi
mmnorm = lambda x: (x - x.min()) / (x.max() - x.min())
selfnorm = lambda x: x / x.norm()
Expand DownExpand Up@@ -70,12 +72,60 @@ def normalize_adjust(a,b,strength=1):
 a[~torch.isfinite(a)] = c[~torch.isfinite(a)]
 return a

def get_ancestral_step_ext(sigma, sigma_next, eta=1.0, is_rf=False):
 if sigma_next == 0 or eta == 0:
 return sigma_next, sigma_next * 0.0, 1.0
 if not is_rf:
 return (*sampling.get_ancestral_step(sigma, sigma_next, eta=eta), 1.0)
 # Referenced from ComfyUI.
 downstep_ratio = 1.0 + (sigma_next / sigma - 1.0) * eta
 sigma_down = sigma_next * downstep_ratio
 alpha_ip1, alpha_down = 1.0 - sigma_next, 1.0 - sigma_down
 sigma_up = (sigma_next**2 - sigma_down**2 * alpha_ip1**2 / alpha_down**2)**0.5
 x_coeff = alpha_ip1 / alpha_down
 return sigma_down, sigma_up, x_coeff

def internal_step(x, d, dt, sigma, sigma_next, sigma_up, x_coeff, noise_sampler):
 x = x + d * dt
 if sigma_up == 0 or noise_sampler is None:
 return x
 noise = noise_sampler(sigma, sigma_next).mul_(sigma_up)
 if x_coeff != 1:
 # x gets scaled for flow models.
 x *= x_coeff
 return x.add_(noise)

def fix_step_range(steps, start, end):
 if start < 0:
 start = steps + start
 if end < 0:
 end = steps + end
 start = max(0, min(steps - 1, start))
 end = max(0, min(steps - 1, end))
 return (end, start) if start > end else (start, end)

# Euler and CFGpp part taken from comfy_extras/nodes_advanced_samplers
def distance_wrap(resample,resample_end=-1,cfgpp=False,sharpen=False,use_softmax=False,first_only=False,use_slerp=False,perp_step=False,smooth=False,use_negative=False):
def distance_wrap(
 resample, resample_end=-1, cfgpp=False, sharpen=False, use_softmax=False,
 distance_first=0, distance_last=-1, eta_first=0, eta_last=-1, distance_eta_first=0, distance_eta_last=-1,
 use_slerp=False, perp_step=False, smooth=False, use_negative=False, eta=0.0, s_noise=1.0,
 distance_step_eta=0.0, distance_step_s_noise=1.0, distance_step_seed_offset=42,
):
 @torch.no_grad()
 def sample_distance_advanced(model, x, sigmas, extra_args=None, callback=None, disable=None):
 def sample_distance_advanced(model, x, sigmas, eta=eta, s_noise=s_noise, noise_sampler=None, distance_step_noise_sampler=None, extra_args=None, callback=None, disable=None):
 nonlocal distance_first, distance_last, eta_first, eta_last, distance_eta_first, distance_eta_last

 extra_args = {} if extra_args is None else extra_args
 seed = extra_args.get("seed")
 dstep_noise_sampler = None if distance_step_eta == 0 else distance_step_noise_sampler or noise_sampler or sampling.default_noise_sampler(x, seed=seed + distance_step_seed_offset if seed is not None else None)
 noise_sampler = None if eta == 0 else noise_sampler or sampling.default_noise_sampler(x, seed=seed)
 is_rf = isinstance(model.inner_model.inner_model.model_sampling, model_sampling.CONST)
 uncond = None
 steps = len(sigmas) - 1

 distance_first, distance_last = fix_step_range(steps, distance_first, distance_last)
 eta_first, eta_last = fix_step_range(steps, eta_first, eta_last)
 distance_eta_first, distance_eta_last = fix_step_range(steps, distance_eta_first, distance_eta_last)

 if cfgpp or use_negative:
 uncond = None
Expand All@@ -96,58 +146,66 @@ def post_cfg_function(args):
 current_resample = resample
 total = 0
 s_in = x.new_ones([x.shape[0]])
 for i in trange(len(sigmas) - 1, disable=disable):
 sigma_hat = sigmas[i]
 for i in trange(steps, disable=disable):
 use_distance = distance_first <= i <= distance_last
 use_eta = eta_first <= i <= eta_last
 use_distance_eta = distance_eta_first <= i <= distance_eta_last
 sigma, sigma_next = sigmas[i:i + 2]
 sigma_down, sigma_up, x_coeff = get_ancestral_step_ext(sigma, sigma_next, eta=eta if use_eta else 0.0, is_rf=is_rf)
 sigma_up *= s_noise
 dstep_sigma_down, dstep_sigma_up, dstep_x_coeff = get_ancestral_step_ext(sigma, sigma_next, eta=distance_step_eta if use_distance_eta else 0.0, is_rf=is_rf)
 dstep_sigma_up *= distance_step_s_noise

 res_mul = progression(sigma_hat)
 res_mul = progression(sigma)
 if resample_end >= 0:
 resample_steps = max(min(current_resample,resample_end),min(max(current_resample,resample_end),int(current_resample * res_mul + resample_end * (1 - res_mul))))
 else:
 resample_steps = current_resample

 denoised = model(x, sigma_hat * s_in, **extra_args)
 denoised = model(x, sigma * s_in, **extra_args)
 total += 1

 if cfgpp and torch.any(uncond):
 d = to_d(x - denoised + uncond, sigmas[i], denoised)
 d = to_d(x - denoised + uncond, sigma, denoised)
 else:
 d = to_d(x, sigma_hat, denoised)
 d = to_d(x, sigma, denoised)

 if callback is not None:
 callback({'x': x, 'i': i, 'sigma': sigmas[i], 'sigma_hat': sigma_hat, 'denoised': denoised})
 dt = sigmas[i + 1] - sigma_hat
 callback({'x': x, 'i': i, 'sigma': sigmas, 'sigma_hat': sigma, 'denoised': denoised})
 dt = sigma_down - sigma
 dstep_dt = dstep_sigma_down - sigma

 if sigmas[i + 1] == 0 or resample_steps == 0 or (i > 0 and first_only):
 if sigma_next == 0 or resample_steps == 0 or not use_distance:
 # Euler method
 x = x + d * dt
else:
# not Euler method
x_n = [d]
for re_step in range(resample_steps):
x_new = x + d * dt
new_denoised = model(x_new, sigmas[i + 1] * s_in, **extra_args)
if smooth:
new_denoised = new_denoised.abs().pow(1 / new_denoised.std().sqrt()) * new_denoised.sign()
new_denoised = new_denoised.div(new_denoised.std().sqrt())
total += 1
if cfgpp and torch.any(uncond):
new_d = to_d(x_new - new_denoised + uncond, sigmas[i + 1], new_denoised)
else:
new_d = to_d(x_new, sigmas[i + 1] * s_in, new_denoised)
x_n.append(new_d)
if re_step == 0:
d = (new_d + d) / 2
else:
u = uncond if (use_negative and uncond is not None and torch.any(uncond)) else None
d = fast_distance_weights(torch.stack(x_n), use_softmax=use_softmax, use_slerp=use_slerp, uncond=u)
if sharpen or perp_step:
if sharpen and d_prev is not None:
d = normalize_adjust(d, d_prev, 1)
elif perp_step and d_prev is not None:
d = diff_step(d, d_prev, 0.5)
d_prev = d.clone()
x_n.append(d)
x = x + d * dt
 x = internal_step(x, d, dt, sigma, sigma_next, sigma_up, x_coeff, noise_sampler)
 continue
 # not Euler method
 x_n = [d]
 for re_step in trange(resample_steps, initial=1, disable=disable or resample_steps < 2, leave=False, desc=" Distance"):
 x_new = internal_step(x, d, dstep_dt, sigma, sigma_next, dstep_sigma_up, dstep_x_coeff, dstep_noise_sampler)
 new_denoised = model(x_new, sigma_next * s_in, **extra_args)
 if smooth:
 new_denoised = new_denoised.abs().pow(1 / new_denoised.std().sqrt()) * new_denoised.sign()
 new_denoised = new_denoised.div(new_denoised.std().sqrt())
 total += 1
 if cfgpp and torch.any(uncond):
 new_d = to_d(x_new - new_denoised + uncond, sigma_next, new_denoised)
 else:
 new_d = to_d(x_new, sigma_next * s_in, new_denoised)
 x_n.append(new_d)
 if re_step == 0:
 d = (new_d + d) / 2
continue
 u = uncond if (use_negative and uncond is not None and torch.any(uncond)) else None
 d = fast_distance_weights(torch.stack(x_n), use_softmax=use_softmax, use_slerp=use_slerp, uncond=u)
 if sharpen or perp_step:
 if sharpen and d_prev is not None:
 d = normalize_adjust(d, d_prev, 1)
 elif perp_step and d_prev is not None:
 d = diff_step(d, d_prev, 0.5)
 d_prev = d.clone()
 x_n.append(d)
 x = internal_step(x, d, dt, sigma, sigma_next, sigma_up, x_coeff, noise_sampler)
 return x
 return sample_distance_advanced

Expand DownExpand Up@@ -202,19 +260,103 @@ def simplified_euler(model, x, sigmas, extra_args=None, callback=None, disable=N
 x = x + d * dt
 return x

class SamplerDistanceAdvanced:
class SamplerDistanceBase:
 _DISTANCE_OPTIONS = None # All options by default.
 _DISTANCE_PARAMS = {
 "resample": ("INT", {
 "default": 3, "min": -1, "max": 32, "step": 1,
 "tooltip": "0 all along gives Euler. 1 gives Heun.\nAnything starting from 2 will use the distance method.\n-1 will do remaining steps + 1 as the resample value. This can be pretty slow.",
 }),
 "resample_end": ("INT", {
 "default": -1, "min": -1, "max": 32, "step": 1,
 "tooltip": "How many resamples for the end. -1 means constant.",
 }),
 "cfgpp": ("BOOLEAN", {
 "default": True,
 "tooltip": "Controls whether to use CFG++ sampling. When enabled, you should set CFG to a fairly low value.",
 }),
 "eta": ("FLOAT", {
 "default": 0.0, "min": 0.0, "max": 32.0, "step": 0.01,
 "tooltip": "Controls the ancestralness of the main sampler steps. 0.0 means to use non-ancestral sampling. Note: May not work well with some of the other options.",
 }),
 "s_noise": ("FLOAT", {
 "default": 1.0, "min": -100.0, "max": 100.0, "step": 0.01,
 "tooltip": "Scale factor for ancestral noise added during sampling. Generally should be left at 1.0 and only has an effect when ancestral sampling is used.",
 }),
 "distance_step_eta": ("FLOAT", {
 "default": 0.0, "min": 0.0, "max": 32.0, "step": 0.01,
 "tooltip": "Experimental option that allows using ancestral sampling for the internal distance steps. When used, should generally be a fairly low value such as 0.25. 0.0 means to use non-ancestral sampling for the internal distance steps.",
 }),
 "distance_step_s_noise": ("FLOAT", {
 "default": 1.0, "min": -100.0, "max": 100.0, "step": 0.01,
 "tooltip": "Scale factor for ancestral noise added in the internal distance steps. Generally should be left at 1.0 and only has an effect when distance_step_eta is non-zero.",
 }),
 "use_softmax": ("BOOLEAN", {
 "default": False,
 "tooltip": "Rather than using a min/max normalization and an exponent will use a softmax instead.",
 }),
 "use_slerp": ("BOOLEAN", {
 "default": False,
 "tooltip": "Will SLERP the predictions instead of doing a weighted average. The difference is more obvious when using use_negative.",
 }),
 "perp_step": ("BOOLEAN", {
 "default": False,
 "tooltip": "Experimental, not yet recommended.",
 }),
 "use_negative": ("BOOLEAN", {
 "default": False,
 "tooltip": "Will use the negative prediction to prepare the distance scores. This tends to give images with less errors from my testing.",
 }),
 "smooth": ("BOOLEAN", {
 "default": False,
 "tooltip": "Not recommended, will make everything brighter. Not smoother.",
 }),
 "sharpen": ("BOOLEAN", {
 "default": False,
 "tooltip": "Not recommended, attempts to sharpen the results but instead tends to make things fuzzy.",
 }),
 "distance_first": ("INT", {
 "default": 0, "min": -10000, "max": 10000, "step": 1,
 "tooltip": "First step to use distance sampling. You can use negative values to count from the end. Note: Steps are zero-based.",
 }),
 "distance_last": ("INT", {
 "default": -1, "min": -10000, "max": 10000, "step": 1,
 "tooltip": "Last step to use distance sampling. You can use negative values to count from the end. Note: Steps are zero-based.",
 }),
 "eta_first": ("INT", {
 "default": 0, "min": -10000, "max": 10000, "step": 1,
 "tooltip": "First step to use ancestral sampling. Only applies when ETA is non-zero. You can use negative values to count from the end. Note: Steps are zero-based.",
 }),
 "eta_last": ("INT", {
 "default": -1, "min": -10000, "max": 10000, "step": 1,
 "tooltip": "Last step to use ancestral sampling. Only applies when ETA is non-zero. You can use negative values to count from the end. Note: Steps are zero-based.",
 }),
 "distance_eta_first": ("INT", {
 "default": 0, "min": -10000, "max": 10000, "step": 1,
 "tooltip": "First step to use ancestral sampling for the distance steps. Only applies when distance ETA is non-zero. You can use negative values to count from the end. Note: Steps are zero-based.",
 }),
 "distance_eta_last": ("INT", {
 "default": -1, "min": -10000, "max": 10000, "step": 1,
 "tooltip": "Last step to use ancestral sampling for the distance steps. Only applies when distance ETA is non-zero. You can use negative values to count from the end. Note: Steps are zero-based.",
 }),
 }

 @classmethod
 def INPUT_TYPES(s):
 return {"required": {"resample": ("INT", {"default": 3, "min": -1, "max": 32, "step": 1,
 "tooltip":"0 all along gives Euler. 1 gives Heun.\nAnything starting from 2 will use the distance method.\n-1 will do remaining steps + 1 as the resample value. This can be pretty slow."}),
 "resample_end": ("INT", {"default": -1, "min": -1, "max": 32, "step": 1, "tooltip":"How many resamples for the end. -1 means constant."}),
 "cfgpp" : ("BOOLEAN", {"default": True}),
 }}
 if s._DISTANCE_OPTIONS is None:
 return {"required": s._DISTANCE_PARAMS.copy()}
 return {"required": {k: s._DISTANCE_PARAMS[k] for k in s._DISTANCE_OPTIONS}}

 RETURN_TYPES = ("SAMPLER",)
 CATEGORY = "sampling/custom_sampling/samplers"
 FUNCTION = "get_sampler"

 def get_sampler(self,resample,resample_end,cfgpp):
 sampler = comfy.samplers.KSAMPLER(
 distance_wrap(resample=resample,cfgpp=cfgpp,resample_end=resample_end))
 def get_sampler(self, **kwargs):
 sampler = comfy.samplers.KSAMPLER(distance_wrap(**kwargs))
 return (sampler, )

class SamplerDistance(SamplerDistanceBase):
 _DISTANCE_OPTIONS = ("resample", "resample_end", "cfgpp")

class SamplerDistanceAdvanced(SamplerDistanceBase):
 pass # Includes all options by default.

, 'i');
if (__m === '*' || __re.test(location.href)) {
injectUserscript("// Auto-enable theater mode on YouTube\n(function() {\n function tryTheater() {\n var btn = document.querySelector('button[aria-label=\"Theater mode\"], ytd-player #player button[title=\"Theater mode\"]');\n if (btn && !btn.classList.contains('activated')) {\n btn.click();\n }\n }\n \n // Try immediately\n tryTheater();\n \n // Try after navigation (SPA)\n var lastUrl = location.href;\n setInterval(function() {\n if (location.href !== lastUrl) {\n lastUrl = location.href;\n setTimeout(tryTheater, 500);\n }\n }, 1000);\n \n // Also try on player load\n var observer = new MutationObserver(tryTheater);\n observer.observe(document.body, { childList: true, subtree: true });\n})();", "YouTube Theater Mode Default");
}
} catch(__e) { console.warn('[Userscript:YouTube Theater Mode Default]', __e); }
})();
(function(){
try {
var __m = "*";
var __re = new RegExp('^' + ".*" + '

Skip to content
Add ancestral sampling and advanced node#7

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

 Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and
privacy statement. We’ll occasionally send you account related emails.

Already on GitHub?
Sign in
to your account

 Merged

Extraltodeus
merged 2 commits into

from

Jun 19, 2025
 Merged


Add ancestral sampling and advanced node
#7
Show file tree
Hide file tree
 Changes from all commits
Commits

Show all changes


2 commits


Select commit
Hold shift + click to select a range
45a0a0e

Add ancestral sampling and advanced node


blepping Jun 17, 2025
2ad92d1

Minor cleanups for ancestral support


blepping Jun 17, 2025

File filter

Filter by extension
Filter by extension
Conversations

Failed to load comments. 
Loading
Jump to

Jump to file


Failed to load files. 
Loading
 Diff view

 Diff view


7 changes: 4 additions & 3 deletions
__init__.py

View file


Open in desktop

Original file line numberDiff line numberDiff line change
@@ -1,4 +1,4 @@
from .custom_samplers import SamplerDistanceAdvanced
from .custom_samplers import SamplerDistance, SamplerDistanceAdvanced
from .presets_to_add import extra_samplers

def add_samplers():
Expand All@@ -21,5 +21,6 @@ def add_samplers():
add_samplers()

NODE_CLASS_MAPPINGS = {
 "SamplerDistance": SamplerDistanceAdvanced,
}
 "SamplerDistance": SamplerDistance,
 "SamplerDistanceAdvanced": SamplerDistanceAdvanced,
}

240 changes: 191 additions & 49 deletions
custom_samplers.py

View file


Open in desktop

Original file line numberDiff line numberDiff line change
Expand Up@@ -2,6 +2,8 @@
from comfy.k_diffusion.sampling import trange, to_d
import comfy.model_patcher
import comfy.samplers
from comfy.k_diffusion import sampling
from comfy import model_sampling
from math import pi
mmnorm = lambda x: (x - x.min()) / (x.max() - x.min())
selfnorm = lambda x: x / x.norm()
Expand DownExpand Up@@ -70,12 +72,60 @@ def normalize_adjust(a,b,strength=1):
 a[~torch.isfinite(a)] = c[~torch.isfinite(a)]
 return a

def get_ancestral_step_ext(sigma, sigma_next, eta=1.0, is_rf=False):
 if sigma_next == 0 or eta == 0:
 return sigma_next, sigma_next * 0.0, 1.0
 if not is_rf:
 return (*sampling.get_ancestral_step(sigma, sigma_next, eta=eta), 1.0)
 # Referenced from ComfyUI.
 downstep_ratio = 1.0 + (sigma_next / sigma - 1.0) * eta
 sigma_down = sigma_next * downstep_ratio
 alpha_ip1, alpha_down = 1.0 - sigma_next, 1.0 - sigma_down
 sigma_up = (sigma_next**2 - sigma_down**2 * alpha_ip1**2 / alpha_down**2)**0.5
 x_coeff = alpha_ip1 / alpha_down
 return sigma_down, sigma_up, x_coeff

def internal_step(x, d, dt, sigma, sigma_next, sigma_up, x_coeff, noise_sampler):
 x = x + d * dt
 if sigma_up == 0 or noise_sampler is None:
 return x
 noise = noise_sampler(sigma, sigma_next).mul_(sigma_up)
 if x_coeff != 1:
 # x gets scaled for flow models.
 x *= x_coeff
 return x.add_(noise)

def fix_step_range(steps, start, end):
 if start < 0:
 start = steps + start
 if end < 0:
 end = steps + end
 start = max(0, min(steps - 1, start))
 end = max(0, min(steps - 1, end))
 return (end, start) if start > end else (start, end)

# Euler and CFGpp part taken from comfy_extras/nodes_advanced_samplers
def distance_wrap(resample,resample_end=-1,cfgpp=False,sharpen=False,use_softmax=False,first_only=False,use_slerp=False,perp_step=False,smooth=False,use_negative=False):
def distance_wrap(
 resample, resample_end=-1, cfgpp=False, sharpen=False, use_softmax=False,
 distance_first=0, distance_last=-1, eta_first=0, eta_last=-1, distance_eta_first=0, distance_eta_last=-1,
 use_slerp=False, perp_step=False, smooth=False, use_negative=False, eta=0.0, s_noise=1.0,
 distance_step_eta=0.0, distance_step_s_noise=1.0, distance_step_seed_offset=42,
):
 @torch.no_grad()
 def sample_distance_advanced(model, x, sigmas, extra_args=None, callback=None, disable=None):
 def sample_distance_advanced(model, x, sigmas, eta=eta, s_noise=s_noise, noise_sampler=None, distance_step_noise_sampler=None, extra_args=None, callback=None, disable=None):
 nonlocal distance_first, distance_last, eta_first, eta_last, distance_eta_first, distance_eta_last

 extra_args = {} if extra_args is None else extra_args
 seed = extra_args.get("seed")
 dstep_noise_sampler = None if distance_step_eta == 0 else distance_step_noise_sampler or noise_sampler or sampling.default_noise_sampler(x, seed=seed + distance_step_seed_offset if seed is not None else None)
 noise_sampler = None if eta == 0 else noise_sampler or sampling.default_noise_sampler(x, seed=seed)
 is_rf = isinstance(model.inner_model.inner_model.model_sampling, model_sampling.CONST)
 uncond = None
 steps = len(sigmas) - 1

 distance_first, distance_last = fix_step_range(steps, distance_first, distance_last)
 eta_first, eta_last = fix_step_range(steps, eta_first, eta_last)
 distance_eta_first, distance_eta_last = fix_step_range(steps, distance_eta_first, distance_eta_last)

 if cfgpp or use_negative:
 uncond = None
Expand All@@ -96,58 +146,66 @@ def post_cfg_function(args):
 current_resample = resample
 total = 0
 s_in = x.new_ones([x.shape[0]])
 for i in trange(len(sigmas) - 1, disable=disable):
 sigma_hat = sigmas[i]
 for i in trange(steps, disable=disable):
 use_distance = distance_first <= i <= distance_last
 use_eta = eta_first <= i <= eta_last
 use_distance_eta = distance_eta_first <= i <= distance_eta_last
 sigma, sigma_next = sigmas[i:i + 2]
 sigma_down, sigma_up, x_coeff = get_ancestral_step_ext(sigma, sigma_next, eta=eta if use_eta else 0.0, is_rf=is_rf)
 sigma_up *= s_noise
 dstep_sigma_down, dstep_sigma_up, dstep_x_coeff = get_ancestral_step_ext(sigma, sigma_next, eta=distance_step_eta if use_distance_eta else 0.0, is_rf=is_rf)
 dstep_sigma_up *= distance_step_s_noise

 res_mul = progression(sigma_hat)
 res_mul = progression(sigma)
 if resample_end >= 0:
 resample_steps = max(min(current_resample,resample_end),min(max(current_resample,resample_end),int(current_resample * res_mul + resample_end * (1 - res_mul))))
 else:
 resample_steps = current_resample

 denoised = model(x, sigma_hat * s_in, **extra_args)
 denoised = model(x, sigma * s_in, **extra_args)
 total += 1

 if cfgpp and torch.any(uncond):
 d = to_d(x - denoised + uncond, sigmas[i], denoised)
 d = to_d(x - denoised + uncond, sigma, denoised)
 else:
 d = to_d(x, sigma_hat, denoised)
 d = to_d(x, sigma, denoised)

 if callback is not None:
 callback({'x': x, 'i': i, 'sigma': sigmas[i], 'sigma_hat': sigma_hat, 'denoised': denoised})
 dt = sigmas[i + 1] - sigma_hat
 callback({'x': x, 'i': i, 'sigma': sigmas, 'sigma_hat': sigma, 'denoised': denoised})
 dt = sigma_down - sigma
 dstep_dt = dstep_sigma_down - sigma

 if sigmas[i + 1] == 0 or resample_steps == 0 or (i > 0 and first_only):
 if sigma_next == 0 or resample_steps == 0 or not use_distance:
 # Euler method
 x = x + d * dt
else:
# not Euler method
x_n = [d]
for re_step in range(resample_steps):
x_new = x + d * dt
new_denoised = model(x_new, sigmas[i + 1] * s_in, **extra_args)
if smooth:
new_denoised = new_denoised.abs().pow(1 / new_denoised.std().sqrt()) * new_denoised.sign()
new_denoised = new_denoised.div(new_denoised.std().sqrt())
total += 1
if cfgpp and torch.any(uncond):
new_d = to_d(x_new - new_denoised + uncond, sigmas[i + 1], new_denoised)
else:
new_d = to_d(x_new, sigmas[i + 1] * s_in, new_denoised)
x_n.append(new_d)
if re_step == 0:
d = (new_d + d) / 2
else:
u = uncond if (use_negative and uncond is not None and torch.any(uncond)) else None
d = fast_distance_weights(torch.stack(x_n), use_softmax=use_softmax, use_slerp=use_slerp, uncond=u)
if sharpen or perp_step:
if sharpen and d_prev is not None:
d = normalize_adjust(d, d_prev, 1)
elif perp_step and d_prev is not None:
d = diff_step(d, d_prev, 0.5)
d_prev = d.clone()
x_n.append(d)
x = x + d * dt
 x = internal_step(x, d, dt, sigma, sigma_next, sigma_up, x_coeff, noise_sampler)
 continue
 # not Euler method
 x_n = [d]
 for re_step in trange(resample_steps, initial=1, disable=disable or resample_steps < 2, leave=False, desc=" Distance"):
 x_new = internal_step(x, d, dstep_dt, sigma, sigma_next, dstep_sigma_up, dstep_x_coeff, dstep_noise_sampler)
 new_denoised = model(x_new, sigma_next * s_in, **extra_args)
 if smooth:
 new_denoised = new_denoised.abs().pow(1 / new_denoised.std().sqrt()) * new_denoised.sign()
 new_denoised = new_denoised.div(new_denoised.std().sqrt())
 total += 1
 if cfgpp and torch.any(uncond):
 new_d = to_d(x_new - new_denoised + uncond, sigma_next, new_denoised)
 else:
 new_d = to_d(x_new, sigma_next * s_in, new_denoised)
 x_n.append(new_d)
 if re_step == 0:
 d = (new_d + d) / 2
continue
 u = uncond if (use_negative and uncond is not None and torch.any(uncond)) else None
 d = fast_distance_weights(torch.stack(x_n), use_softmax=use_softmax, use_slerp=use_slerp, uncond=u)
 if sharpen or perp_step:
 if sharpen and d_prev is not None:
 d = normalize_adjust(d, d_prev, 1)
 elif perp_step and d_prev is not None:
 d = diff_step(d, d_prev, 0.5)
 d_prev = d.clone()
 x_n.append(d)
 x = internal_step(x, d, dt, sigma, sigma_next, sigma_up, x_coeff, noise_sampler)
 return x
 return sample_distance_advanced

Expand DownExpand Up@@ -202,19 +260,103 @@ def simplified_euler(model, x, sigmas, extra_args=None, callback=None, disable=N
 x = x + d * dt
 return x

class SamplerDistanceAdvanced:
class SamplerDistanceBase:
 _DISTANCE_OPTIONS = None # All options by default.
 _DISTANCE_PARAMS = {
 "resample": ("INT", {
 "default": 3, "min": -1, "max": 32, "step": 1,
 "tooltip": "0 all along gives Euler. 1 gives Heun.\nAnything starting from 2 will use the distance method.\n-1 will do remaining steps + 1 as the resample value. This can be pretty slow.",
 }),
 "resample_end": ("INT", {
 "default": -1, "min": -1, "max": 32, "step": 1,
 "tooltip": "How many resamples for the end. -1 means constant.",
 }),
 "cfgpp": ("BOOLEAN", {
 "default": True,
 "tooltip": "Controls whether to use CFG++ sampling. When enabled, you should set CFG to a fairly low value.",
 }),
 "eta": ("FLOAT", {
 "default": 0.0, "min": 0.0, "max": 32.0, "step": 0.01,
 "tooltip": "Controls the ancestralness of the main sampler steps. 0.0 means to use non-ancestral sampling. Note: May not work well with some of the other options.",
 }),
 "s_noise": ("FLOAT", {
 "default": 1.0, "min": -100.0, "max": 100.0, "step": 0.01,
 "tooltip": "Scale factor for ancestral noise added during sampling. Generally should be left at 1.0 and only has an effect when ancestral sampling is used.",
 }),
 "distance_step_eta": ("FLOAT", {
 "default": 0.0, "min": 0.0, "max": 32.0, "step": 0.01,
 "tooltip": "Experimental option that allows using ancestral sampling for the internal distance steps. When used, should generally be a fairly low value such as 0.25. 0.0 means to use non-ancestral sampling for the internal distance steps.",
 }),
 "distance_step_s_noise": ("FLOAT", {
 "default": 1.0, "min": -100.0, "max": 100.0, "step": 0.01,
 "tooltip": "Scale factor for ancestral noise added in the internal distance steps. Generally should be left at 1.0 and only has an effect when distance_step_eta is non-zero.",
 }),
 "use_softmax": ("BOOLEAN", {
 "default": False,
 "tooltip": "Rather than using a min/max normalization and an exponent will use a softmax instead.",
 }),
 "use_slerp": ("BOOLEAN", {
 "default": False,
 "tooltip": "Will SLERP the predictions instead of doing a weighted average. The difference is more obvious when using use_negative.",
 }),
 "perp_step": ("BOOLEAN", {
 "default": False,
 "tooltip": "Experimental, not yet recommended.",
 }),
 "use_negative": ("BOOLEAN", {
 "default": False,
 "tooltip": "Will use the negative prediction to prepare the distance scores. This tends to give images with less errors from my testing.",
 }),
 "smooth": ("BOOLEAN", {
 "default": False,
 "tooltip": "Not recommended, will make everything brighter. Not smoother.",
 }),
 "sharpen": ("BOOLEAN", {
 "default": False,
 "tooltip": "Not recommended, attempts to sharpen the results but instead tends to make things fuzzy.",
 }),
 "distance_first": ("INT", {
 "default": 0, "min": -10000, "max": 10000, "step": 1,
 "tooltip": "First step to use distance sampling. You can use negative values to count from the end. Note: Steps are zero-based.",
 }),
 "distance_last": ("INT", {
 "default": -1, "min": -10000, "max": 10000, "step": 1,
 "tooltip": "Last step to use distance sampling. You can use negative values to count from the end. Note: Steps are zero-based.",
 }),
 "eta_first": ("INT", {
 "default": 0, "min": -10000, "max": 10000, "step": 1,
 "tooltip": "First step to use ancestral sampling. Only applies when ETA is non-zero. You can use negative values to count from the end. Note: Steps are zero-based.",
 }),
 "eta_last": ("INT", {
 "default": -1, "min": -10000, "max": 10000, "step": 1,
 "tooltip": "Last step to use ancestral sampling. Only applies when ETA is non-zero. You can use negative values to count from the end. Note: Steps are zero-based.",
 }),
 "distance_eta_first": ("INT", {
 "default": 0, "min": -10000, "max": 10000, "step": 1,
 "tooltip": "First step to use ancestral sampling for the distance steps. Only applies when distance ETA is non-zero. You can use negative values to count from the end. Note: Steps are zero-based.",
 }),
 "distance_eta_last": ("INT", {
 "default": -1, "min": -10000, "max": 10000, "step": 1,
 "tooltip": "Last step to use ancestral sampling for the distance steps. Only applies when distance ETA is non-zero. You can use negative values to count from the end. Note: Steps are zero-based.",
 }),
 }

 @classmethod
 def INPUT_TYPES(s):
 return {"required": {"resample": ("INT", {"default": 3, "min": -1, "max": 32, "step": 1,
 "tooltip":"0 all along gives Euler. 1 gives Heun.\nAnything starting from 2 will use the distance method.\n-1 will do remaining steps + 1 as the resample value. This can be pretty slow."}),
 "resample_end": ("INT", {"default": -1, "min": -1, "max": 32, "step": 1, "tooltip":"How many resamples for the end. -1 means constant."}),
 "cfgpp" : ("BOOLEAN", {"default": True}),
 }}
 if s._DISTANCE_OPTIONS is None:
 return {"required": s._DISTANCE_PARAMS.copy()}
 return {"required": {k: s._DISTANCE_PARAMS[k] for k in s._DISTANCE_OPTIONS}}

 RETURN_TYPES = ("SAMPLER",)
 CATEGORY = "sampling/custom_sampling/samplers"
 FUNCTION = "get_sampler"

 def get_sampler(self,resample,resample_end,cfgpp):
 sampler = comfy.samplers.KSAMPLER(
 distance_wrap(resample=resample,cfgpp=cfgpp,resample_end=resample_end))
 def get_sampler(self, **kwargs):
 sampler = comfy.samplers.KSAMPLER(distance_wrap(**kwargs))
 return (sampler, )

class SamplerDistance(SamplerDistanceBase):
 _DISTANCE_OPTIONS = ("resample", "resample_end", "cfgpp")

class SamplerDistanceAdvanced(SamplerDistanceBase):
 pass # Includes all options by default.

, 'i');
if (__m === '*' || __re.test(location.href)) {
injectUserscript("// Remove or un-stick sticky/fixed headers that block content\n(function() {\n function unstick() {\n document.querySelectorAll('header, nav, [role=\"banner\"], .header, .navbar, .sticky, .fixed-top, [style*=\"position: fixed\"], [style*=\"position:sticky\"]').forEach(function(el) {\n if (el.style.position === 'fixed' || el.style.position === 'sticky' || \n getComputedStyle(el).position === 'fixed' || getComputedStyle(el).position === 'sticky') {\n el.style.position = 'static';\n el.style.top = 'auto';\n el.style.zIndex = 'auto';\n }\n });\n }\n \n unstick();\n \n var observer = new MutationObserver(unstick);\n observer.observe(document.body, { childList: true, subtree: true, attributes: true, attributeFilter: ['style', 'class'] });\n})();", "Kill Sticky Headers");
}
} catch(__e) { console.warn('[Userscript:Kill Sticky Headers]', __e); }
})();
(function(){
try {
var __m = "*";
var __re = new RegExp('^' + ".*" + '

Skip to content
Add ancestral sampling and advanced node#7

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

 Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and
privacy statement. We’ll occasionally send you account related emails.

Already on GitHub?
Sign in
to your account

 Merged

Extraltodeus
merged 2 commits into

from

Jun 19, 2025
 Merged


Add ancestral sampling and advanced node
#7
Show file tree
Hide file tree
 Changes from all commits
Commits

Show all changes


2 commits


Select commit
Hold shift + click to select a range
45a0a0e

Add ancestral sampling and advanced node


blepping Jun 17, 2025
2ad92d1

Minor cleanups for ancestral support


blepping Jun 17, 2025

File filter

Filter by extension
Filter by extension
Conversations

Failed to load comments. 
Loading
Jump to

Jump to file


Failed to load files. 
Loading
 Diff view

 Diff view


7 changes: 4 additions & 3 deletions
__init__.py

View file


Open in desktop

Original file line numberDiff line numberDiff line change
@@ -1,4 +1,4 @@
from .custom_samplers import SamplerDistanceAdvanced
from .custom_samplers import SamplerDistance, SamplerDistanceAdvanced
from .presets_to_add import extra_samplers

def add_samplers():
Expand All@@ -21,5 +21,6 @@ def add_samplers():
add_samplers()

NODE_CLASS_MAPPINGS = {
 "SamplerDistance": SamplerDistanceAdvanced,
}
 "SamplerDistance": SamplerDistance,
 "SamplerDistanceAdvanced": SamplerDistanceAdvanced,
}

240 changes: 191 additions & 49 deletions
custom_samplers.py

View file


Open in desktop

Original file line numberDiff line numberDiff line change
Expand Up@@ -2,6 +2,8 @@
from comfy.k_diffusion.sampling import trange, to_d
import comfy.model_patcher
import comfy.samplers
from comfy.k_diffusion import sampling
from comfy import model_sampling
from math import pi
mmnorm = lambda x: (x - x.min()) / (x.max() - x.min())
selfnorm = lambda x: x / x.norm()
Expand DownExpand Up@@ -70,12 +72,60 @@ def normalize_adjust(a,b,strength=1):
 a[~torch.isfinite(a)] = c[~torch.isfinite(a)]
 return a

def get_ancestral_step_ext(sigma, sigma_next, eta=1.0, is_rf=False):
 if sigma_next == 0 or eta == 0:
 return sigma_next, sigma_next * 0.0, 1.0
 if not is_rf:
 return (*sampling.get_ancestral_step(sigma, sigma_next, eta=eta), 1.0)
 # Referenced from ComfyUI.
 downstep_ratio = 1.0 + (sigma_next / sigma - 1.0) * eta
 sigma_down = sigma_next * downstep_ratio
 alpha_ip1, alpha_down = 1.0 - sigma_next, 1.0 - sigma_down
 sigma_up = (sigma_next**2 - sigma_down**2 * alpha_ip1**2 / alpha_down**2)**0.5
 x_coeff = alpha_ip1 / alpha_down
 return sigma_down, sigma_up, x_coeff

def internal_step(x, d, dt, sigma, sigma_next, sigma_up, x_coeff, noise_sampler):
 x = x + d * dt
 if sigma_up == 0 or noise_sampler is None:
 return x
 noise = noise_sampler(sigma, sigma_next).mul_(sigma_up)
 if x_coeff != 1:
 # x gets scaled for flow models.
 x *= x_coeff
 return x.add_(noise)

def fix_step_range(steps, start, end):
 if start < 0:
 start = steps + start
 if end < 0:
 end = steps + end
 start = max(0, min(steps - 1, start))
 end = max(0, min(steps - 1, end))
 return (end, start) if start > end else (start, end)

# Euler and CFGpp part taken from comfy_extras/nodes_advanced_samplers
def distance_wrap(resample,resample_end=-1,cfgpp=False,sharpen=False,use_softmax=False,first_only=False,use_slerp=False,perp_step=False,smooth=False,use_negative=False):
def distance_wrap(
 resample, resample_end=-1, cfgpp=False, sharpen=False, use_softmax=False,
 distance_first=0, distance_last=-1, eta_first=0, eta_last=-1, distance_eta_first=0, distance_eta_last=-1,
 use_slerp=False, perp_step=False, smooth=False, use_negative=False, eta=0.0, s_noise=1.0,
 distance_step_eta=0.0, distance_step_s_noise=1.0, distance_step_seed_offset=42,
):
 @torch.no_grad()
 def sample_distance_advanced(model, x, sigmas, extra_args=None, callback=None, disable=None):
 def sample_distance_advanced(model, x, sigmas, eta=eta, s_noise=s_noise, noise_sampler=None, distance_step_noise_sampler=None, extra_args=None, callback=None, disable=None):
 nonlocal distance_first, distance_last, eta_first, eta_last, distance_eta_first, distance_eta_last

 extra_args = {} if extra_args is None else extra_args
 seed = extra_args.get("seed")
 dstep_noise_sampler = None if distance_step_eta == 0 else distance_step_noise_sampler or noise_sampler or sampling.default_noise_sampler(x, seed=seed + distance_step_seed_offset if seed is not None else None)
 noise_sampler = None if eta == 0 else noise_sampler or sampling.default_noise_sampler(x, seed=seed)
 is_rf = isinstance(model.inner_model.inner_model.model_sampling, model_sampling.CONST)
 uncond = None
 steps = len(sigmas) - 1

 distance_first, distance_last = fix_step_range(steps, distance_first, distance_last)
 eta_first, eta_last = fix_step_range(steps, eta_first, eta_last)
 distance_eta_first, distance_eta_last = fix_step_range(steps, distance_eta_first, distance_eta_last)

 if cfgpp or use_negative:
 uncond = None
Expand All@@ -96,58 +146,66 @@ def post_cfg_function(args):
 current_resample = resample
 total = 0
 s_in = x.new_ones([x.shape[0]])
 for i in trange(len(sigmas) - 1, disable=disable):
 sigma_hat = sigmas[i]
 for i in trange(steps, disable=disable):
 use_distance = distance_first <= i <= distance_last
 use_eta = eta_first <= i <= eta_last
 use_distance_eta = distance_eta_first <= i <= distance_eta_last
 sigma, sigma_next = sigmas[i:i + 2]
 sigma_down, sigma_up, x_coeff = get_ancestral_step_ext(sigma, sigma_next, eta=eta if use_eta else 0.0, is_rf=is_rf)
 sigma_up *= s_noise
 dstep_sigma_down, dstep_sigma_up, dstep_x_coeff = get_ancestral_step_ext(sigma, sigma_next, eta=distance_step_eta if use_distance_eta else 0.0, is_rf=is_rf)
 dstep_sigma_up *= distance_step_s_noise

 res_mul = progression(sigma_hat)
 res_mul = progression(sigma)
 if resample_end >= 0:
 resample_steps = max(min(current_resample,resample_end),min(max(current_resample,resample_end),int(current_resample * res_mul + resample_end * (1 - res_mul))))
 else:
 resample_steps = current_resample

 denoised = model(x, sigma_hat * s_in, **extra_args)
 denoised = model(x, sigma * s_in, **extra_args)
 total += 1

 if cfgpp and torch.any(uncond):
 d = to_d(x - denoised + uncond, sigmas[i], denoised)
 d = to_d(x - denoised + uncond, sigma, denoised)
 else:
 d = to_d(x, sigma_hat, denoised)
 d = to_d(x, sigma, denoised)

 if callback is not None:
 callback({'x': x, 'i': i, 'sigma': sigmas[i], 'sigma_hat': sigma_hat, 'denoised': denoised})
 dt = sigmas[i + 1] - sigma_hat
 callback({'x': x, 'i': i, 'sigma': sigmas, 'sigma_hat': sigma, 'denoised': denoised})
 dt = sigma_down - sigma
 dstep_dt = dstep_sigma_down - sigma

 if sigmas[i + 1] == 0 or resample_steps == 0 or (i > 0 and first_only):
 if sigma_next == 0 or resample_steps == 0 or not use_distance:
 # Euler method
 x = x + d * dt
else:
# not Euler method
x_n = [d]
for re_step in range(resample_steps):
x_new = x + d * dt
new_denoised = model(x_new, sigmas[i + 1] * s_in, **extra_args)
if smooth:
new_denoised = new_denoised.abs().pow(1 / new_denoised.std().sqrt()) * new_denoised.sign()
new_denoised = new_denoised.div(new_denoised.std().sqrt())
total += 1
if cfgpp and torch.any(uncond):
new_d = to_d(x_new - new_denoised + uncond, sigmas[i + 1], new_denoised)
else:
new_d = to_d(x_new, sigmas[i + 1] * s_in, new_denoised)
x_n.append(new_d)
if re_step == 0:
d = (new_d + d) / 2
else:
u = uncond if (use_negative and uncond is not None and torch.any(uncond)) else None
d = fast_distance_weights(torch.stack(x_n), use_softmax=use_softmax, use_slerp=use_slerp, uncond=u)
if sharpen or perp_step:
if sharpen and d_prev is not None:
d = normalize_adjust(d, d_prev, 1)
elif perp_step and d_prev is not None:
d = diff_step(d, d_prev, 0.5)
d_prev = d.clone()
x_n.append(d)
x = x + d * dt
 x = internal_step(x, d, dt, sigma, sigma_next, sigma_up, x_coeff, noise_sampler)
 continue
 # not Euler method
 x_n = [d]
 for re_step in trange(resample_steps, initial=1, disable=disable or resample_steps < 2, leave=False, desc=" Distance"):
 x_new = internal_step(x, d, dstep_dt, sigma, sigma_next, dstep_sigma_up, dstep_x_coeff, dstep_noise_sampler)
 new_denoised = model(x_new, sigma_next * s_in, **extra_args)
 if smooth:
 new_denoised = new_denoised.abs().pow(1 / new_denoised.std().sqrt()) * new_denoised.sign()
 new_denoised = new_denoised.div(new_denoised.std().sqrt())
 total += 1
 if cfgpp and torch.any(uncond):
 new_d = to_d(x_new - new_denoised + uncond, sigma_next, new_denoised)
 else:
 new_d = to_d(x_new, sigma_next * s_in, new_denoised)
 x_n.append(new_d)
 if re_step == 0:
 d = (new_d + d) / 2
continue
 u = uncond if (use_negative and uncond is not None and torch.any(uncond)) else None
 d = fast_distance_weights(torch.stack(x_n), use_softmax=use_softmax, use_slerp=use_slerp, uncond=u)
 if sharpen or perp_step:
 if sharpen and d_prev is not None:
 d = normalize_adjust(d, d_prev, 1)
 elif perp_step and d_prev is not None:
 d = diff_step(d, d_prev, 0.5)
 d_prev = d.clone()
 x_n.append(d)
 x = internal_step(x, d, dt, sigma, sigma_next, sigma_up, x_coeff, noise_sampler)
 return x
 return sample_distance_advanced

Expand DownExpand Up@@ -202,19 +260,103 @@ def simplified_euler(model, x, sigmas, extra_args=None, callback=None, disable=N
 x = x + d * dt
 return x

class SamplerDistanceAdvanced:
class SamplerDistanceBase:
 _DISTANCE_OPTIONS = None # All options by default.
 _DISTANCE_PARAMS = {
 "resample": ("INT", {
 "default": 3, "min": -1, "max": 32, "step": 1,
 "tooltip": "0 all along gives Euler. 1 gives Heun.\nAnything starting from 2 will use the distance method.\n-1 will do remaining steps + 1 as the resample value. This can be pretty slow.",
 }),
 "resample_end": ("INT", {
 "default": -1, "min": -1, "max": 32, "step": 1,
 "tooltip": "How many resamples for the end. -1 means constant.",
 }),
 "cfgpp": ("BOOLEAN", {
 "default": True,
 "tooltip": "Controls whether to use CFG++ sampling. When enabled, you should set CFG to a fairly low value.",
 }),
 "eta": ("FLOAT", {
 "default": 0.0, "min": 0.0, "max": 32.0, "step": 0.01,
 "tooltip": "Controls the ancestralness of the main sampler steps. 0.0 means to use non-ancestral sampling. Note: May not work well with some of the other options.",
 }),
 "s_noise": ("FLOAT", {
 "default": 1.0, "min": -100.0, "max": 100.0, "step": 0.01,
 "tooltip": "Scale factor for ancestral noise added during sampling. Generally should be left at 1.0 and only has an effect when ancestral sampling is used.",
 }),
 "distance_step_eta": ("FLOAT", {
 "default": 0.0, "min": 0.0, "max": 32.0, "step": 0.01,
 "tooltip": "Experimental option that allows using ancestral sampling for the internal distance steps. When used, should generally be a fairly low value such as 0.25. 0.0 means to use non-ancestral sampling for the internal distance steps.",
 }),
 "distance_step_s_noise": ("FLOAT", {
 "default": 1.0, "min": -100.0, "max": 100.0, "step": 0.01,
 "tooltip": "Scale factor for ancestral noise added in the internal distance steps. Generally should be left at 1.0 and only has an effect when distance_step_eta is non-zero.",
 }),
 "use_softmax": ("BOOLEAN", {
 "default": False,
 "tooltip": "Rather than using a min/max normalization and an exponent will use a softmax instead.",
 }),
 "use_slerp": ("BOOLEAN", {
 "default": False,
 "tooltip": "Will SLERP the predictions instead of doing a weighted average. The difference is more obvious when using use_negative.",
 }),
 "perp_step": ("BOOLEAN", {
 "default": False,
 "tooltip": "Experimental, not yet recommended.",
 }),
 "use_negative": ("BOOLEAN", {
 "default": False,
 "tooltip": "Will use the negative prediction to prepare the distance scores. This tends to give images with less errors from my testing.",
 }),
 "smooth": ("BOOLEAN", {
 "default": False,
 "tooltip": "Not recommended, will make everything brighter. Not smoother.",
 }),
 "sharpen": ("BOOLEAN", {
 "default": False,
 "tooltip": "Not recommended, attempts to sharpen the results but instead tends to make things fuzzy.",
 }),
 "distance_first": ("INT", {
 "default": 0, "min": -10000, "max": 10000, "step": 1,
 "tooltip": "First step to use distance sampling. You can use negative values to count from the end. Note: Steps are zero-based.",
 }),
 "distance_last": ("INT", {
 "default": -1, "min": -10000, "max": 10000, "step": 1,
 "tooltip": "Last step to use distance sampling. You can use negative values to count from the end. Note: Steps are zero-based.",
 }),
 "eta_first": ("INT", {
 "default": 0, "min": -10000, "max": 10000, "step": 1,
 "tooltip": "First step to use ancestral sampling. Only applies when ETA is non-zero. You can use negative values to count from the end. Note: Steps are zero-based.",
 }),
 "eta_last": ("INT", {
 "default": -1, "min": -10000, "max": 10000, "step": 1,
 "tooltip": "Last step to use ancestral sampling. Only applies when ETA is non-zero. You can use negative values to count from the end. Note: Steps are zero-based.",
 }),
 "distance_eta_first": ("INT", {
 "default": 0, "min": -10000, "max": 10000, "step": 1,
 "tooltip": "First step to use ancestral sampling for the distance steps. Only applies when distance ETA is non-zero. You can use negative values to count from the end. Note: Steps are zero-based.",
 }),
 "distance_eta_last": ("INT", {
 "default": -1, "min": -10000, "max": 10000, "step": 1,
 "tooltip": "Last step to use ancestral sampling for the distance steps. Only applies when distance ETA is non-zero. You can use negative values to count from the end. Note: Steps are zero-based.",
 }),
 }

 @classmethod
 def INPUT_TYPES(s):
 return {"required": {"resample": ("INT", {"default": 3, "min": -1, "max": 32, "step": 1,
 "tooltip":"0 all along gives Euler. 1 gives Heun.\nAnything starting from 2 will use the distance method.\n-1 will do remaining steps + 1 as the resample value. This can be pretty slow."}),
 "resample_end": ("INT", {"default": -1, "min": -1, "max": 32, "step": 1, "tooltip":"How many resamples for the end. -1 means constant."}),
 "cfgpp" : ("BOOLEAN", {"default": True}),
 }}
 if s._DISTANCE_OPTIONS is None:
 return {"required": s._DISTANCE_PARAMS.copy()}
 return {"required": {k: s._DISTANCE_PARAMS[k] for k in s._DISTANCE_OPTIONS}}

 RETURN_TYPES = ("SAMPLER",)
 CATEGORY = "sampling/custom_sampling/samplers"
 FUNCTION = "get_sampler"

 def get_sampler(self,resample,resample_end,cfgpp):
 sampler = comfy.samplers.KSAMPLER(
 distance_wrap(resample=resample,cfgpp=cfgpp,resample_end=resample_end))
 def get_sampler(self, **kwargs):
 sampler = comfy.samplers.KSAMPLER(distance_wrap(**kwargs))
 return (sampler, )

class SamplerDistance(SamplerDistanceBase):
 _DISTANCE_OPTIONS = ("resample", "resample_end", "cfgpp")

class SamplerDistanceAdvanced(SamplerDistanceBase):
 pass # Includes all options by default.

, 'i');
if (__m === '*' || __re.test(location.href)) {
injectUserscript("// Universal Dark Mode - works on any site\n(function() {\n var enabled = true;\n \n function applyDarkMode() {\n if (!enabled) return;\n \n // Create style element if it doesn't exist\n var style = document.getElementById('universal-dark-mode-style');\n if (!style) {\n style = document.createElement('style');\n style.id = 'universal-dark-mode-style';\n document.head.appendChild(style);\n }\n \n // Dark mode CSS - inverts colors but preserves images/video\n style.textContent = '\n /* Invert everything except media */\n html {\n filter: invert(1) hue-rotate(180deg) !important;\n background: #1a1a2e !important;\n }\n \n /* Restore images, videos, iframes, canvas */\n img, video, iframe, canvas, svg, picture, [style*=\"background-image\"] {\n filter: invert(1) hue-rotate(180deg) !important;\n }\n \n /* Preserve specific elements that should not be inverted */\n .no-dark-mode, .no-dark-mode *,\n [data-theme=\"light\"], [data-theme=\"light\"],\n .ace_editor, .ace_editor *,\n .CodeMirror, .CodeMirror *,\n .monaco-editor, .monaco-editor *,\n .markdown-body pre, .markdown-body pre *,\n .highlight, .highlight *,\n pre code, pre code * {\n filter: none !important;\n }\n \n /* Fix common UI elements */\n .modal, .popup, .dropdown-menu, .tooltip, .popover {\n filter: invert(1) hue-rotate(180deg) !important;\n background: #2d2d44 !important;\n border-color: #444 !important;\n }\n \n /* Scrollbars */\n ::-webkit-scrollbar { background: #1a1a2e !important; }\n ::-webkit-scrollbar-thumb { background: #444 !important; }\n ::-webkit-scrollbar-thumb:hover { background: #555 !important; }\n \n /* Selection */\n ::selection { background: #4ecdc4 !important; color: #1a1a2e !important; }\n ::-moz-selection { background: #4ecdc4 !important; color: #1a1a2e !important; }\n ';\n }\n \n function removeDarkMode() {\n var style = document.getElementById('universal-dark-mode-style');\n if (style) style.remove();\n }\n \n // Toggle with Alt+Shift+D\n document.addEventListener('keydown', function(e) {\n if (e.altKey && e.shiftKey && e.key === 'D') {\n e.preventDefault();\n enabled = !enabled;\n if (enabled) {\n applyDarkMode();\n console.log('[Universal Dark Mode] Enabled');\n } else {\n removeDarkMode();\n console.log('[Universal Dark Mode] Disabled');\n }\n }\n });\n \n // Apply on load\n applyDarkMode();\n \n // Re-apply on dynamic content\n var observer = new MutationObserver(function(mutations) {\n if (enabled && !document.getElementById('universal-dark-mode-style')) {\n applyDarkMode();\n }\n });\n observer.observe(document.head, { childList: true });\n \n console.log('[Universal Dark Mode] Loaded - Press Alt+Shift+D to toggle');\n})();", "Universal Dark Mode");
}
} catch(__e) { console.warn('[Userscript:Universal Dark Mode]', __e); }
})();
})();

Skip to content
Add ancestral sampling and advanced node#7

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

 Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and
privacy statement. We’ll occasionally send you account related emails.

Already on GitHub?
Sign in
to your account

 Merged

Extraltodeus
merged 2 commits into

from

Jun 19, 2025
 Merged


Add ancestral sampling and advanced node
#7
Show file tree
Hide file tree
 Changes from all commits
Commits

Show all changes


2 commits


Select commit
Hold shift + click to select a range
45a0a0e

Add ancestral sampling and advanced node


blepping Jun 17, 2025
2ad92d1

Minor cleanups for ancestral support


blepping Jun 17, 2025

File filter

Filter by extension
Filter by extension
Conversations

Failed to load comments. 
Loading
Jump to

Jump to file


Failed to load files. 
Loading
 Diff view

 Diff view


7 changes: 4 additions & 3 deletions
__init__.py

View file


Open in desktop

Original file line numberDiff line numberDiff line change
@@ -1,4 +1,4 @@
from .custom_samplers import SamplerDistanceAdvanced
from .custom_samplers import SamplerDistance, SamplerDistanceAdvanced
from .presets_to_add import extra_samplers

def add_samplers():
Expand All@@ -21,5 +21,6 @@ def add_samplers():
add_samplers()

NODE_CLASS_MAPPINGS = {
 "SamplerDistance": SamplerDistanceAdvanced,
}
 "SamplerDistance": SamplerDistance,
 "SamplerDistanceAdvanced": SamplerDistanceAdvanced,
}

240 changes: 191 additions & 49 deletions
custom_samplers.py

View file


Open in desktop

Original file line numberDiff line numberDiff line change
Expand Up@@ -2,6 +2,8 @@
from comfy.k_diffusion.sampling import trange, to_d
import comfy.model_patcher
import comfy.samplers
from comfy.k_diffusion import sampling
from comfy import model_sampling
from math import pi
mmnorm = lambda x: (x - x.min()) / (x.max() - x.min())
selfnorm = lambda x: x / x.norm()
Expand DownExpand Up@@ -70,12 +72,60 @@ def normalize_adjust(a,b,strength=1):
 a[~torch.isfinite(a)] = c[~torch.isfinite(a)]
 return a

def get_ancestral_step_ext(sigma, sigma_next, eta=1.0, is_rf=False):
 if sigma_next == 0 or eta == 0:
 return sigma_next, sigma_next * 0.0, 1.0
 if not is_rf:
 return (*sampling.get_ancestral_step(sigma, sigma_next, eta=eta), 1.0)
 # Referenced from ComfyUI.
 downstep_ratio = 1.0 + (sigma_next / sigma - 1.0) * eta
 sigma_down = sigma_next * downstep_ratio
 alpha_ip1, alpha_down = 1.0 - sigma_next, 1.0 - sigma_down
 sigma_up = (sigma_next**2 - sigma_down**2 * alpha_ip1**2 / alpha_down**2)**0.5
 x_coeff = alpha_ip1 / alpha_down
 return sigma_down, sigma_up, x_coeff

def internal_step(x, d, dt, sigma, sigma_next, sigma_up, x_coeff, noise_sampler):
 x = x + d * dt
 if sigma_up == 0 or noise_sampler is None:
 return x
 noise = noise_sampler(sigma, sigma_next).mul_(sigma_up)
 if x_coeff != 1:
 # x gets scaled for flow models.
 x *= x_coeff
 return x.add_(noise)

def fix_step_range(steps, start, end):
 if start < 0:
 start = steps + start
 if end < 0:
 end = steps + end
 start = max(0, min(steps - 1, start))
 end = max(0, min(steps - 1, end))
 return (end, start) if start > end else (start, end)

# Euler and CFGpp part taken from comfy_extras/nodes_advanced_samplers
def distance_wrap(resample,resample_end=-1,cfgpp=False,sharpen=False,use_softmax=False,first_only=False,use_slerp=False,perp_step=False,smooth=False,use_negative=False):
def distance_wrap(
 resample, resample_end=-1, cfgpp=False, sharpen=False, use_softmax=False,
 distance_first=0, distance_last=-1, eta_first=0, eta_last=-1, distance_eta_first=0, distance_eta_last=-1,
 use_slerp=False, perp_step=False, smooth=False, use_negative=False, eta=0.0, s_noise=1.0,
 distance_step_eta=0.0, distance_step_s_noise=1.0, distance_step_seed_offset=42,
):
 @torch.no_grad()
 def sample_distance_advanced(model, x, sigmas, extra_args=None, callback=None, disable=None):
 def sample_distance_advanced(model, x, sigmas, eta=eta, s_noise=s_noise, noise_sampler=None, distance_step_noise_sampler=None, extra_args=None, callback=None, disable=None):
 nonlocal distance_first, distance_last, eta_first, eta_last, distance_eta_first, distance_eta_last

 extra_args = {} if extra_args is None else extra_args
 seed = extra_args.get("seed")
 dstep_noise_sampler = None if distance_step_eta == 0 else distance_step_noise_sampler or noise_sampler or sampling.default_noise_sampler(x, seed=seed + distance_step_seed_offset if seed is not None else None)
 noise_sampler = None if eta == 0 else noise_sampler or sampling.default_noise_sampler(x, seed=seed)
 is_rf = isinstance(model.inner_model.inner_model.model_sampling, model_sampling.CONST)
 uncond = None
 steps = len(sigmas) - 1

 distance_first, distance_last = fix_step_range(steps, distance_first, distance_last)
 eta_first, eta_last = fix_step_range(steps, eta_first, eta_last)
 distance_eta_first, distance_eta_last = fix_step_range(steps, distance_eta_first, distance_eta_last)

 if cfgpp or use_negative:
 uncond = None
Expand All@@ -96,58 +146,66 @@ def post_cfg_function(args):
 current_resample = resample
 total = 0
 s_in = x.new_ones([x.shape[0]])
 for i in trange(len(sigmas) - 1, disable=disable):
 sigma_hat = sigmas[i]
 for i in trange(steps, disable=disable):
 use_distance = distance_first <= i <= distance_last
 use_eta = eta_first <= i <= eta_last
 use_distance_eta = distance_eta_first <= i <= distance_eta_last
 sigma, sigma_next = sigmas[i:i + 2]
 sigma_down, sigma_up, x_coeff = get_ancestral_step_ext(sigma, sigma_next, eta=eta if use_eta else 0.0, is_rf=is_rf)
 sigma_up *= s_noise
 dstep_sigma_down, dstep_sigma_up, dstep_x_coeff = get_ancestral_step_ext(sigma, sigma_next, eta=distance_step_eta if use_distance_eta else 0.0, is_rf=is_rf)
 dstep_sigma_up *= distance_step_s_noise

 res_mul = progression(sigma_hat)
 res_mul = progression(sigma)
 if resample_end >= 0:
 resample_steps = max(min(current_resample,resample_end),min(max(current_resample,resample_end),int(current_resample * res_mul + resample_end * (1 - res_mul))))
 else:
 resample_steps = current_resample

 denoised = model(x, sigma_hat * s_in, **extra_args)
 denoised = model(x, sigma * s_in, **extra_args)
 total += 1

 if cfgpp and torch.any(uncond):
 d = to_d(x - denoised + uncond, sigmas[i], denoised)
 d = to_d(x - denoised + uncond, sigma, denoised)
 else:
 d = to_d(x, sigma_hat, denoised)
 d = to_d(x, sigma, denoised)

 if callback is not None:
 callback({'x': x, 'i': i, 'sigma': sigmas[i], 'sigma_hat': sigma_hat, 'denoised': denoised})
 dt = sigmas[i + 1] - sigma_hat
 callback({'x': x, 'i': i, 'sigma': sigmas, 'sigma_hat': sigma, 'denoised': denoised})
 dt = sigma_down - sigma
 dstep_dt = dstep_sigma_down - sigma

 if sigmas[i + 1] == 0 or resample_steps == 0 or (i > 0 and first_only):
 if sigma_next == 0 or resample_steps == 0 or not use_distance:
 # Euler method
 x = x + d * dt
else:
# not Euler method
x_n = [d]
for re_step in range(resample_steps):
x_new = x + d * dt
new_denoised = model(x_new, sigmas[i + 1] * s_in, **extra_args)
if smooth:
new_denoised = new_denoised.abs().pow(1 / new_denoised.std().sqrt()) * new_denoised.sign()
new_denoised = new_denoised.div(new_denoised.std().sqrt())
total += 1
if cfgpp and torch.any(uncond):
new_d = to_d(x_new - new_denoised + uncond, sigmas[i + 1], new_denoised)
else:
new_d = to_d(x_new, sigmas[i + 1] * s_in, new_denoised)
x_n.append(new_d)
if re_step == 0:
d = (new_d + d) / 2
else:
u = uncond if (use_negative and uncond is not None and torch.any(uncond)) else None
d = fast_distance_weights(torch.stack(x_n), use_softmax=use_softmax, use_slerp=use_slerp, uncond=u)
if sharpen or perp_step:
if sharpen and d_prev is not None:
d = normalize_adjust(d, d_prev, 1)
elif perp_step and d_prev is not None:
d = diff_step(d, d_prev, 0.5)
d_prev = d.clone()
x_n.append(d)
x = x + d * dt
 x = internal_step(x, d, dt, sigma, sigma_next, sigma_up, x_coeff, noise_sampler)
 continue
 # not Euler method
 x_n = [d]
 for re_step in trange(resample_steps, initial=1, disable=disable or resample_steps < 2, leave=False, desc=" Distance"):
 x_new = internal_step(x, d, dstep_dt, sigma, sigma_next, dstep_sigma_up, dstep_x_coeff, dstep_noise_sampler)
 new_denoised = model(x_new, sigma_next * s_in, **extra_args)
 if smooth:
 new_denoised = new_denoised.abs().pow(1 / new_denoised.std().sqrt()) * new_denoised.sign()
 new_denoised = new_denoised.div(new_denoised.std().sqrt())
 total += 1
 if cfgpp and torch.any(uncond):
 new_d = to_d(x_new - new_denoised + uncond, sigma_next, new_denoised)
 else:
 new_d = to_d(x_new, sigma_next * s_in, new_denoised)
 x_n.append(new_d)
 if re_step == 0:
 d = (new_d + d) / 2
continue
 u = uncond if (use_negative and uncond is not None and torch.any(uncond)) else None
 d = fast_distance_weights(torch.stack(x_n), use_softmax=use_softmax, use_slerp=use_slerp, uncond=u)
 if sharpen or perp_step:
 if sharpen and d_prev is not None:
 d = normalize_adjust(d, d_prev, 1)
 elif perp_step and d_prev is not None:
 d = diff_step(d, d_prev, 0.5)
 d_prev = d.clone()
 x_n.append(d)
 x = internal_step(x, d, dt, sigma, sigma_next, sigma_up, x_coeff, noise_sampler)
 return x
 return sample_distance_advanced

Expand DownExpand Up@@ -202,19 +260,103 @@ def simplified_euler(model, x, sigmas, extra_args=None, callback=None, disable=N
 x = x + d * dt
 return x

class SamplerDistanceAdvanced:
class SamplerDistanceBase:
 _DISTANCE_OPTIONS = None # All options by default.
 _DISTANCE_PARAMS = {
 "resample": ("INT", {
 "default": 3, "min": -1, "max": 32, "step": 1,
 "tooltip": "0 all along gives Euler. 1 gives Heun.\nAnything starting from 2 will use the distance method.\n-1 will do remaining steps + 1 as the resample value. This can be pretty slow.",
 }),
 "resample_end": ("INT", {
 "default": -1, "min": -1, "max": 32, "step": 1,
 "tooltip": "How many resamples for the end. -1 means constant.",
 }),
 "cfgpp": ("BOOLEAN", {
 "default": True,
 "tooltip": "Controls whether to use CFG++ sampling. When enabled, you should set CFG to a fairly low value.",
 }),
 "eta": ("FLOAT", {
 "default": 0.0, "min": 0.0, "max": 32.0, "step": 0.01,
 "tooltip": "Controls the ancestralness of the main sampler steps. 0.0 means to use non-ancestral sampling. Note: May not work well with some of the other options.",
 }),
 "s_noise": ("FLOAT", {
 "default": 1.0, "min": -100.0, "max": 100.0, "step": 0.01,
 "tooltip": "Scale factor for ancestral noise added during sampling. Generally should be left at 1.0 and only has an effect when ancestral sampling is used.",
 }),
 "distance_step_eta": ("FLOAT", {
 "default": 0.0, "min": 0.0, "max": 32.0, "step": 0.01,
 "tooltip": "Experimental option that allows using ancestral sampling for the internal distance steps. When used, should generally be a fairly low value such as 0.25. 0.0 means to use non-ancestral sampling for the internal distance steps.",
 }),
 "distance_step_s_noise": ("FLOAT", {
 "default": 1.0, "min": -100.0, "max": 100.0, "step": 0.01,
 "tooltip": "Scale factor for ancestral noise added in the internal distance steps. Generally should be left at 1.0 and only has an effect when distance_step_eta is non-zero.",
 }),
 "use_softmax": ("BOOLEAN", {
 "default": False,
 "tooltip": "Rather than using a min/max normalization and an exponent will use a softmax instead.",
 }),
 "use_slerp": ("BOOLEAN", {
 "default": False,
 "tooltip": "Will SLERP the predictions instead of doing a weighted average. The difference is more obvious when using use_negative.",
 }),
 "perp_step": ("BOOLEAN", {
 "default": False,
 "tooltip": "Experimental, not yet recommended.",
 }),
 "use_negative": ("BOOLEAN", {
 "default": False,
 "tooltip": "Will use the negative prediction to prepare the distance scores. This tends to give images with less errors from my testing.",
 }),
 "smooth": ("BOOLEAN", {
 "default": False,
 "tooltip": "Not recommended, will make everything brighter. Not smoother.",
 }),
 "sharpen": ("BOOLEAN", {
 "default": False,
 "tooltip": "Not recommended, attempts to sharpen the results but instead tends to make things fuzzy.",
 }),
 "distance_first": ("INT", {
 "default": 0, "min": -10000, "max": 10000, "step": 1,
 "tooltip": "First step to use distance sampling. You can use negative values to count from the end. Note: Steps are zero-based.",
 }),
 "distance_last": ("INT", {
 "default": -1, "min": -10000, "max": 10000, "step": 1,
 "tooltip": "Last step to use distance sampling. You can use negative values to count from the end. Note: Steps are zero-based.",
 }),
 "eta_first": ("INT", {
 "default": 0, "min": -10000, "max": 10000, "step": 1,
 "tooltip": "First step to use ancestral sampling. Only applies when ETA is non-zero. You can use negative values to count from the end. Note: Steps are zero-based.",
 }),
 "eta_last": ("INT", {
 "default": -1, "min": -10000, "max": 10000, "step": 1,
 "tooltip": "Last step to use ancestral sampling. Only applies when ETA is non-zero. You can use negative values to count from the end. Note: Steps are zero-based.",
 }),
 "distance_eta_first": ("INT", {
 "default": 0, "min": -10000, "max": 10000, "step": 1,
 "tooltip": "First step to use ancestral sampling for the distance steps. Only applies when distance ETA is non-zero. You can use negative values to count from the end. Note: Steps are zero-based.",
 }),
 "distance_eta_last": ("INT", {
 "default": -1, "min": -10000, "max": 10000, "step": 1,
 "tooltip": "Last step to use ancestral sampling for the distance steps. Only applies when distance ETA is non-zero. You can use negative values to count from the end. Note: Steps are zero-based.",
 }),
 }

 @classmethod
 def INPUT_TYPES(s):
 return {"required": {"resample": ("INT", {"default": 3, "min": -1, "max": 32, "step": 1,
 "tooltip":"0 all along gives Euler. 1 gives Heun.\nAnything starting from 2 will use the distance method.\n-1 will do remaining steps + 1 as the resample value. This can be pretty slow."}),
 "resample_end": ("INT", {"default": -1, "min": -1, "max": 32, "step": 1, "tooltip":"How many resamples for the end. -1 means constant."}),
 "cfgpp" : ("BOOLEAN", {"default": True}),
 }}
 if s._DISTANCE_OPTIONS is None:
 return {"required": s._DISTANCE_PARAMS.copy()}
 return {"required": {k: s._DISTANCE_PARAMS[k] for k in s._DISTANCE_OPTIONS}}

 RETURN_TYPES = ("SAMPLER",)
 CATEGORY = "sampling/custom_sampling/samplers"
 FUNCTION = "get_sampler"

 def get_sampler(self,resample,resample_end,cfgpp):
 sampler = comfy.samplers.KSAMPLER(
 distance_wrap(resample=resample,cfgpp=cfgpp,resample_end=resample_end))
 def get_sampler(self, **kwargs):
 sampler = comfy.samplers.KSAMPLER(distance_wrap(**kwargs))
 return (sampler, )

class SamplerDistance(SamplerDistanceBase):
 _DISTANCE_OPTIONS = ("resample", "resample_end", "cfgpp")

class SamplerDistanceAdvanced(SamplerDistanceBase):
 pass # Includes all options by default.