nodes.py 73.3 KB
Newer Older
comfyanonymous's avatar
comfyanonymous committed
1
2
3
4
5
import torch

import os
import sys
import json
6
import hashlib
7
import traceback
8
import math
9
import time
10
import random
11
import logging
comfyanonymous's avatar
comfyanonymous committed
12

13
from PIL import Image, ImageOps, ImageSequence
comfyanonymous's avatar
comfyanonymous committed
14
15
from PIL.PngImagePlugin import PngInfo
import numpy as np
16
import safetensors.torch
comfyanonymous's avatar
comfyanonymous committed
17

18
19
20
sys.path.insert(0, os.path.join(os.path.dirname(os.path.realpath(__file__)), "comfy"))


21
import comfy.diffusers_load
comfyanonymous's avatar
comfyanonymous committed
22
import comfy.samplers
23
import comfy.sample
comfyanonymous's avatar
comfyanonymous committed
24
import comfy.sd
comfyanonymous's avatar
comfyanonymous committed
25
import comfy.utils
26
import comfy.controlnet
comfyanonymous's avatar
comfyanonymous committed
27

28
import comfy.clip_vision
29

30
import comfy.model_management
31
32
from comfy.cli_args import args

33
import importlib
comfyanonymous's avatar
comfyanonymous committed
34

35
import folder_paths
36
import latent_preview
37
import node_helpers
space-nuko's avatar
space-nuko committed
38

39
def before_node_execution():
40
    comfy.model_management.throw_exception_if_processing_interrupted()
41

42
def interrupt_processing(value=True):
43
    comfy.model_management.interrupt_current_processing(value)
44

comfyanonymous's avatar
comfyanonymous committed
45
MAX_RESOLUTION=16384
46

comfyanonymous's avatar
comfyanonymous committed
47
48
49
class CLIPTextEncode:
    @classmethod
    def INPUT_TYPES(s):
50
        return {"required": {"text": ("STRING", {"multiline": True, "dynamicPrompts": True}), "clip": ("CLIP", )}}
comfyanonymous's avatar
comfyanonymous committed
51
52
53
    RETURN_TYPES = ("CONDITIONING",)
    FUNCTION = "encode"

54
55
    CATEGORY = "conditioning"

comfyanonymous's avatar
comfyanonymous committed
56
    def encode(self, clip, text):
57
58
59
        tokens = clip.tokenize(text)
        cond, pooled = clip.encode_from_tokens(tokens, return_pooled=True)
        return ([[cond, {"pooled_output": pooled}]], )
comfyanonymous's avatar
comfyanonymous committed
60
61
62
63
64
65
66
67

class ConditioningCombine:
    @classmethod
    def INPUT_TYPES(s):
        return {"required": {"conditioning_1": ("CONDITIONING", ), "conditioning_2": ("CONDITIONING", )}}
    RETURN_TYPES = ("CONDITIONING",)
    FUNCTION = "combine"

68
69
    CATEGORY = "conditioning"

comfyanonymous's avatar
comfyanonymous committed
70
71
72
    def combine(self, conditioning_1, conditioning_2):
        return (conditioning_1 + conditioning_2, )

FizzleDorf's avatar
FizzleDorf committed
73
74
75
class ConditioningAverage :
    @classmethod
    def INPUT_TYPES(s):
comfyanonymous's avatar
comfyanonymous committed
76
77
        return {"required": {"conditioning_to": ("CONDITIONING", ), "conditioning_from": ("CONDITIONING", ),
                              "conditioning_to_strength": ("FLOAT", {"default": 1.0, "min": 0.0, "max": 1.0, "step": 0.01})
FizzleDorf's avatar
FizzleDorf committed
78
79
80
81
82
83
                             }}
    RETURN_TYPES = ("CONDITIONING",)
    FUNCTION = "addWeighted"

    CATEGORY = "conditioning"

comfyanonymous's avatar
comfyanonymous committed
84
    def addWeighted(self, conditioning_to, conditioning_from, conditioning_to_strength):
FizzleDorf's avatar
FizzleDorf committed
85
        out = []
comfyanonymous's avatar
comfyanonymous committed
86
87

        if len(conditioning_from) > 1:
88
            logging.warning("Warning: ConditioningAverage conditioning_from contains more than 1 cond, only the first one will actually be applied to conditioning_to.")
comfyanonymous's avatar
comfyanonymous committed
89
90

        cond_from = conditioning_from[0][0]
91
        pooled_output_from = conditioning_from[0][1].get("pooled_output", None)
comfyanonymous's avatar
comfyanonymous committed
92
93
94

        for i in range(len(conditioning_to)):
            t1 = conditioning_to[i][0]
95
            pooled_output_to = conditioning_to[i][1].get("pooled_output", pooled_output_from)
comfyanonymous's avatar
comfyanonymous committed
96
97
98
99
100
            t0 = cond_from[:,:t1.shape[1]]
            if t0.shape[1] < t1.shape[1]:
                t0 = torch.cat([t0] + [torch.zeros((1, (t1.shape[1] - t0.shape[1]), t1.shape[2]))], dim=1)

            tw = torch.mul(t1, conditioning_to_strength) + torch.mul(t0, (1.0 - conditioning_to_strength))
101
102
103
104
105
106
107
            t_to = conditioning_to[i][1].copy()
            if pooled_output_from is not None and pooled_output_to is not None:
                t_to["pooled_output"] = torch.mul(pooled_output_to, conditioning_to_strength) + torch.mul(pooled_output_from, (1.0 - conditioning_to_strength))
            elif pooled_output_from is not None:
                t_to["pooled_output"] = pooled_output_from

            n = [tw, t_to]
FizzleDorf's avatar
FizzleDorf committed
108
109
110
            out.append(n)
        return (out, )

111
112
113
114
115
116
117
118
119
120
class ConditioningConcat:
    @classmethod
    def INPUT_TYPES(s):
        return {"required": {
            "conditioning_to": ("CONDITIONING",),
            "conditioning_from": ("CONDITIONING",),
            }}
    RETURN_TYPES = ("CONDITIONING",)
    FUNCTION = "concat"

121
    CATEGORY = "conditioning"
122
123
124
125
126

    def concat(self, conditioning_to, conditioning_from):
        out = []

        if len(conditioning_from) > 1:
127
            logging.warning("Warning: ConditioningConcat conditioning_from contains more than 1 cond, only the first one will actually be applied to conditioning_to.")
128
129
130
131
132
133
134
135
136
137
138

        cond_from = conditioning_from[0][0]

        for i in range(len(conditioning_to)):
            t1 = conditioning_to[i][0]
            tw = torch.cat((t1, cond_from),1)
            n = [tw, conditioning_to[i][1].copy()]
            out.append(n)

        return (out, )

comfyanonymous's avatar
comfyanonymous committed
139
140
141
142
class ConditioningSetArea:
    @classmethod
    def INPUT_TYPES(s):
        return {"required": {"conditioning": ("CONDITIONING", ),
143
144
145
146
                              "width": ("INT", {"default": 64, "min": 64, "max": MAX_RESOLUTION, "step": 8}),
                              "height": ("INT", {"default": 64, "min": 64, "max": MAX_RESOLUTION, "step": 8}),
                              "x": ("INT", {"default": 0, "min": 0, "max": MAX_RESOLUTION, "step": 8}),
                              "y": ("INT", {"default": 0, "min": 0, "max": MAX_RESOLUTION, "step": 8}),
comfyanonymous's avatar
comfyanonymous committed
147
148
149
150
151
                              "strength": ("FLOAT", {"default": 1.0, "min": 0.0, "max": 10.0, "step": 0.01}),
                             }}
    RETURN_TYPES = ("CONDITIONING",)
    FUNCTION = "append"

152
153
    CATEGORY = "conditioning"

154
    def append(self, conditioning, width, height, x, y, strength):
155
156
157
        c = node_helpers.conditioning_set_values(conditioning, {"area": (height // 8, width // 8, y // 8, x // 8),
                                                                "strength": strength,
                                                                "set_area_to_bounds": False})
comfyanonymous's avatar
comfyanonymous committed
158
        return (c, )
comfyanonymous's avatar
comfyanonymous committed
159

160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
class ConditioningSetAreaPercentage:
    @classmethod
    def INPUT_TYPES(s):
        return {"required": {"conditioning": ("CONDITIONING", ),
                              "width": ("FLOAT", {"default": 1.0, "min": 0, "max": 1.0, "step": 0.01}),
                              "height": ("FLOAT", {"default": 1.0, "min": 0, "max": 1.0, "step": 0.01}),
                              "x": ("FLOAT", {"default": 0, "min": 0, "max": 1.0, "step": 0.01}),
                              "y": ("FLOAT", {"default": 0, "min": 0, "max": 1.0, "step": 0.01}),
                              "strength": ("FLOAT", {"default": 1.0, "min": 0.0, "max": 10.0, "step": 0.01}),
                             }}
    RETURN_TYPES = ("CONDITIONING",)
    FUNCTION = "append"

    CATEGORY = "conditioning"

    def append(self, conditioning, width, height, x, y, strength):
176
177
178
        c = node_helpers.conditioning_set_values(conditioning, {"area": ("percentage", height, width, y, x),
                                                                "strength": strength,
                                                                "set_area_to_bounds": False})
179
180
        return (c, )

181
182
183
184
185
186
187
188
189
190
191
192
class ConditioningSetAreaStrength:
    @classmethod
    def INPUT_TYPES(s):
        return {"required": {"conditioning": ("CONDITIONING", ),
                              "strength": ("FLOAT", {"default": 1.0, "min": 0.0, "max": 10.0, "step": 0.01}),
                             }}
    RETURN_TYPES = ("CONDITIONING",)
    FUNCTION = "append"

    CATEGORY = "conditioning"

    def append(self, conditioning, strength):
193
        c = node_helpers.conditioning_set_values(conditioning, {"strength": strength})
194
195
196
        return (c, )


Jacob Segal's avatar
Jacob Segal committed
197
198
199
200
201
202
class ConditioningSetMask:
    @classmethod
    def INPUT_TYPES(s):
        return {"required": {"conditioning": ("CONDITIONING", ),
                              "mask": ("MASK", ),
                              "strength": ("FLOAT", {"default": 1.0, "min": 0.0, "max": 10.0, "step": 0.01}),
203
                              "set_cond_area": (["default", "mask bounds"],),
Jacob Segal's avatar
Jacob Segal committed
204
205
206
207
208
209
                             }}
    RETURN_TYPES = ("CONDITIONING",)
    FUNCTION = "append"

    CATEGORY = "conditioning"

210
211
212
213
    def append(self, conditioning, mask, set_cond_area, strength):
        set_area_to_bounds = False
        if set_cond_area != "default":
            set_area_to_bounds = True
Jacob Segal's avatar
Jacob Segal committed
214
215
        if len(mask.shape) < 3:
            mask = mask.unsqueeze(0)
216
217
218
219

        c = node_helpers.conditioning_set_values(conditioning, {"mask": mask,
                                                                "set_area_to_bounds": set_area_to_bounds,
                                                                "mask_strength": strength})
Jacob Segal's avatar
Jacob Segal committed
220
221
        return (c, )

222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
class ConditioningZeroOut:
    @classmethod
    def INPUT_TYPES(s):
        return {"required": {"conditioning": ("CONDITIONING", )}}
    RETURN_TYPES = ("CONDITIONING",)
    FUNCTION = "zero_out"

    CATEGORY = "advanced/conditioning"

    def zero_out(self, conditioning):
        c = []
        for t in conditioning:
            d = t[1].copy()
            if "pooled_output" in d:
                d["pooled_output"] = torch.zeros_like(d["pooled_output"])
            n = [torch.zeros_like(t[0]), d]
            c.append(n)
        return (c, )

241
242
243
244
class ConditioningSetTimestepRange:
    @classmethod
    def INPUT_TYPES(s):
        return {"required": {"conditioning": ("CONDITIONING", ),
245
246
                             "start": ("FLOAT", {"default": 0.0, "min": 0.0, "max": 1.0, "step": 0.001}),
                             "end": ("FLOAT", {"default": 1.0, "min": 0.0, "max": 1.0, "step": 0.001})
247
248
249
250
251
252
253
                             }}
    RETURN_TYPES = ("CONDITIONING",)
    FUNCTION = "set_range"

    CATEGORY = "advanced/conditioning"

    def set_range(self, conditioning, start, end):
254
255
        c = node_helpers.conditioning_set_values(conditioning, {"start_percent": start,
                                                                "end_percent": end})
256
257
        return (c, )

comfyanonymous's avatar
comfyanonymous committed
258
259
260
261
262
263
264
class VAEDecode:
    @classmethod
    def INPUT_TYPES(s):
        return {"required": { "samples": ("LATENT", ), "vae": ("VAE", )}}
    RETURN_TYPES = ("IMAGE",)
    FUNCTION = "decode"

265
266
    CATEGORY = "latent"

comfyanonymous's avatar
comfyanonymous committed
267
    def decode(self, vae, samples):
268
        return (vae.decode(samples["samples"]), )
comfyanonymous's avatar
comfyanonymous committed
269

270
271
272
class VAEDecodeTiled:
    @classmethod
    def INPUT_TYPES(s):
273
        return {"required": {"samples": ("LATENT", ), "vae": ("VAE", ),
comfyanonymous's avatar
comfyanonymous committed
274
                             "tile_size": ("INT", {"default": 512, "min": 320, "max": 4096, "step": 64})
275
                            }}
276
277
278
279
280
    RETURN_TYPES = ("IMAGE",)
    FUNCTION = "decode"

    CATEGORY = "_for_testing"

281
    def decode(self, vae, samples, tile_size):
282
        return (vae.decode_tiled(samples["samples"], tile_x=tile_size // 8, tile_y=tile_size // 8, ), )
283

comfyanonymous's avatar
comfyanonymous committed
284
285
286
287
288
289
290
class VAEEncode:
    @classmethod
    def INPUT_TYPES(s):
        return {"required": { "pixels": ("IMAGE", ), "vae": ("VAE", )}}
    RETURN_TYPES = ("LATENT",)
    FUNCTION = "encode"

291
292
    CATEGORY = "latent"

293
294
    def encode(self, vae, pixels):
        t = vae.encode(pixels[:,:,:,:3])
295
        return ({"samples":t}, )
comfyanonymous's avatar
comfyanonymous committed
296

comfyanonymous's avatar
comfyanonymous committed
297
298
299
class VAEEncodeTiled:
    @classmethod
    def INPUT_TYPES(s):
300
        return {"required": {"pixels": ("IMAGE", ), "vae": ("VAE", ),
301
                             "tile_size": ("INT", {"default": 512, "min": 320, "max": 4096, "step": 64})
302
                            }}
comfyanonymous's avatar
comfyanonymous committed
303
304
305
306
307
    RETURN_TYPES = ("LATENT",)
    FUNCTION = "encode"

    CATEGORY = "_for_testing"

308
309
    def encode(self, vae, pixels, tile_size):
        t = vae.encode_tiled(pixels[:,:,:,:3], tile_x=tile_size, tile_y=tile_size, )
comfyanonymous's avatar
comfyanonymous committed
310
        return ({"samples":t}, )
311

312
313
314
class VAEEncodeForInpaint:
    @classmethod
    def INPUT_TYPES(s):
315
        return {"required": { "pixels": ("IMAGE", ), "vae": ("VAE", ), "mask": ("MASK", ), "grow_mask_by": ("INT", {"default": 6, "min": 0, "max": 64, "step": 1}),}}
316
317
318
319
320
    RETURN_TYPES = ("LATENT",)
    FUNCTION = "encode"

    CATEGORY = "latent/inpaint"

321
    def encode(self, vae, pixels, mask, grow_mask_by=6):
322
323
        x = (pixels.shape[1] // vae.downscale_ratio) * vae.downscale_ratio
        y = (pixels.shape[2] // vae.downscale_ratio) * vae.downscale_ratio
324
        mask = torch.nn.functional.interpolate(mask.reshape((-1, 1, mask.shape[-2], mask.shape[-1])), size=(pixels.shape[1], pixels.shape[2]), mode="bilinear")
325

326
        pixels = pixels.clone()
327
        if pixels.shape[1] != x or pixels.shape[2] != y:
328
329
            x_offset = (pixels.shape[1] % vae.downscale_ratio) // 2
            y_offset = (pixels.shape[2] % vae.downscale_ratio) // 2
330
331
            pixels = pixels[:,x_offset:x + x_offset, y_offset:y + y_offset,:]
            mask = mask[:,:,x_offset:x + x_offset, y_offset:y + y_offset]
332

333
        #grow mask by a few pixels to keep things seamless in latent space
334
335
336
337
338
339
340
341
        if grow_mask_by == 0:
            mask_erosion = mask
        else:
            kernel_tensor = torch.ones((1, 1, grow_mask_by, grow_mask_by))
            padding = math.ceil((grow_mask_by - 1) / 2)

            mask_erosion = torch.clamp(torch.nn.functional.conv2d(mask.round(), kernel_tensor, padding=padding), 0, 1)

342
        m = (1.0 - mask.round()).squeeze(1)
343
344
        for i in range(3):
            pixels[:,:,:,i] -= 0.5
345
            pixels[:,:,:,i] *= m
346
347
348
            pixels[:,:,:,i] += 0.5
        t = vae.encode(pixels)

349
        return ({"samples":t, "noise_mask": (mask_erosion[:,:,:x,:y].round())}, )
comfyanonymous's avatar
comfyanonymous committed
350

351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395

class InpaintModelConditioning:
    @classmethod
    def INPUT_TYPES(s):
        return {"required": {"positive": ("CONDITIONING", ),
                             "negative": ("CONDITIONING", ),
                             "vae": ("VAE", ),
                             "pixels": ("IMAGE", ),
                             "mask": ("MASK", ),
                             }}

    RETURN_TYPES = ("CONDITIONING","CONDITIONING","LATENT")
    RETURN_NAMES = ("positive", "negative", "latent")
    FUNCTION = "encode"

    CATEGORY = "conditioning/inpaint"

    def encode(self, positive, negative, pixels, vae, mask):
        x = (pixels.shape[1] // 8) * 8
        y = (pixels.shape[2] // 8) * 8
        mask = torch.nn.functional.interpolate(mask.reshape((-1, 1, mask.shape[-2], mask.shape[-1])), size=(pixels.shape[1], pixels.shape[2]), mode="bilinear")

        orig_pixels = pixels
        pixels = orig_pixels.clone()
        if pixels.shape[1] != x or pixels.shape[2] != y:
            x_offset = (pixels.shape[1] % 8) // 2
            y_offset = (pixels.shape[2] % 8) // 2
            pixels = pixels[:,x_offset:x + x_offset, y_offset:y + y_offset,:]
            mask = mask[:,:,x_offset:x + x_offset, y_offset:y + y_offset]

        m = (1.0 - mask.round()).squeeze(1)
        for i in range(3):
            pixels[:,:,:,i] -= 0.5
            pixels[:,:,:,i] *= m
            pixels[:,:,:,i] += 0.5
        concat_latent = vae.encode(pixels)
        orig_latent = vae.encode(orig_pixels)

        out_latent = {}

        out_latent["samples"] = orig_latent
        out_latent["noise_mask"] = mask

        out = []
        for conditioning in [positive, negative]:
396
397
            c = node_helpers.conditioning_set_values(conditioning, {"concat_latent_image": concat_latent,
                                                                    "concat_mask": mask})
398
399
400
401
            out.append(c)
        return (out[0], out[1], out_latent)


Dr.Lt.Data's avatar
Dr.Lt.Data committed
402
403
class SaveLatent:
    def __init__(self):
404
        self.output_dir = folder_paths.get_output_directory()
Dr.Lt.Data's avatar
Dr.Lt.Data committed
405
406
407
408

    @classmethod
    def INPUT_TYPES(s):
        return {"required": { "samples": ("LATENT", ),
409
                              "filename_prefix": ("STRING", {"default": "latents/ComfyUI"})},
Dr.Lt.Data's avatar
Dr.Lt.Data committed
410
411
412
413
414
415
416
417
418
419
                "hidden": {"prompt": "PROMPT", "extra_pnginfo": "EXTRA_PNGINFO"},
                }
    RETURN_TYPES = ()
    FUNCTION = "save"

    OUTPUT_NODE = True

    CATEGORY = "_for_testing"

    def save(self, samples, filename_prefix="ComfyUI", prompt=None, extra_pnginfo=None):
420
        full_output_folder, filename, counter, subfolder, filename_prefix = folder_paths.get_save_image_path(filename_prefix, self.output_dir)
Dr.Lt.Data's avatar
Dr.Lt.Data committed
421
422
423
424
425
426

        # support save metadata for latent sharing
        prompt_info = ""
        if prompt is not None:
            prompt_info = json.dumps(prompt)

427
428
429
430
431
432
        metadata = None
        if not args.disable_metadata:
            metadata = {"prompt": prompt_info}
            if extra_pnginfo is not None:
                for x in extra_pnginfo:
                    metadata[x] = json.dumps(extra_pnginfo[x])
Dr.Lt.Data's avatar
Dr.Lt.Data committed
433
434

        file = f"{filename}_{counter:05}_.latent"
435
436
437
438
439
440
441
442

        results = list()
        results.append({
            "filename": file,
            "subfolder": subfolder,
            "type": "output"
        })

Dr.Lt.Data's avatar
Dr.Lt.Data committed
443
444
        file = os.path.join(full_output_folder, file)

445
446
        output = {}
        output["latent_tensor"] = samples["samples"]
447
        output["latent_format_version_0"] = torch.tensor([])
448

449
        comfy.utils.save_torch_file(output, file, metadata=metadata)
450
        return { "ui": { "latents": results } }
Dr.Lt.Data's avatar
Dr.Lt.Data committed
451
452
453
454
455


class LoadLatent:
    @classmethod
    def INPUT_TYPES(s):
456
457
        input_dir = folder_paths.get_input_directory()
        files = [f for f in os.listdir(input_dir) if os.path.isfile(os.path.join(input_dir, f)) and f.endswith(".latent")]
Dr.Lt.Data's avatar
Dr.Lt.Data committed
458
459
460
461
462
463
464
465
        return {"required": {"latent": [sorted(files), ]}, }

    CATEGORY = "_for_testing"

    RETURN_TYPES = ("LATENT", )
    FUNCTION = "load"

    def load(self, latent):
466
467
        latent_path = folder_paths.get_annotated_filepath(latent)
        latent = safetensors.torch.load_file(latent_path, device="cpu")
468
469
470
471
        multiplier = 1.0
        if "latent_format_version_0" not in latent:
            multiplier = 1.0 / 0.18215
        samples = {"samples": latent["latent_tensor"].float() * multiplier}
472
        return (samples, )
Dr.Lt.Data's avatar
Dr.Lt.Data committed
473

474
475
476
477
478
479
480
481
482
483
484
485
486
487
    @classmethod
    def IS_CHANGED(s, latent):
        image_path = folder_paths.get_annotated_filepath(latent)
        m = hashlib.sha256()
        with open(image_path, 'rb') as f:
            m.update(f.read())
        return m.digest().hex()

    @classmethod
    def VALIDATE_INPUTS(s, latent):
        if not folder_paths.exists_annotated_filepath(latent):
            return "Invalid latent file: {}".format(latent)
        return True

Dr.Lt.Data's avatar
Dr.Lt.Data committed
488

comfyanonymous's avatar
comfyanonymous committed
489
490
491
class CheckpointLoader:
    @classmethod
    def INPUT_TYPES(s):
492
493
        return {"required": { "config_name": (folder_paths.get_filename_list("configs"), ),
                              "ckpt_name": (folder_paths.get_filename_list("checkpoints"), )}}
comfyanonymous's avatar
comfyanonymous committed
494
495
496
    RETURN_TYPES = ("MODEL", "CLIP", "VAE")
    FUNCTION = "load_checkpoint"

497
    CATEGORY = "advanced/loaders"
498

comfyanonymous's avatar
comfyanonymous committed
499
    def load_checkpoint(self, config_name, ckpt_name, output_vae=True, output_clip=True):
500
501
        config_path = folder_paths.get_full_path("configs", config_name)
        ckpt_path = folder_paths.get_full_path("checkpoints", ckpt_name)
502
        return comfy.sd.load_checkpoint(config_path, ckpt_path, output_vae=True, output_clip=True, embedding_directory=folder_paths.get_folder_paths("embeddings"))
comfyanonymous's avatar
comfyanonymous committed
503

504
505
506
class CheckpointLoaderSimple:
    @classmethod
    def INPUT_TYPES(s):
507
        return {"required": { "ckpt_name": (folder_paths.get_filename_list("checkpoints"), ),
508
509
510
511
                             }}
    RETURN_TYPES = ("MODEL", "CLIP", "VAE")
    FUNCTION = "load_checkpoint"

512
    CATEGORY = "loaders"
513

514
    def load_checkpoint(self, ckpt_name, output_vae=True, output_clip=True):
515
        ckpt_path = folder_paths.get_full_path("checkpoints", ckpt_name)
516
        out = comfy.sd.load_checkpoint_guess_config(ckpt_path, output_vae=True, output_clip=True, embedding_directory=folder_paths.get_folder_paths("embeddings"))
517
        return out[:3]
518

sALTaccount's avatar
sALTaccount committed
519
520
521
class DiffusersLoader:
    @classmethod
    def INPUT_TYPES(cls):
522
        paths = []
sALTaccount's avatar
sALTaccount committed
523
        for search_path in folder_paths.get_folder_paths("diffusers"):
524
            if os.path.exists(search_path):
525
526
527
528
                for root, subdir, files in os.walk(search_path, followlinks=True):
                    if "model_index.json" in files:
                        paths.append(os.path.relpath(root, start=search_path))

529
        return {"required": {"model_path": (paths,), }}
sALTaccount's avatar
sALTaccount committed
530
531
532
    RETURN_TYPES = ("MODEL", "CLIP", "VAE")
    FUNCTION = "load_checkpoint"

533
    CATEGORY = "advanced/loaders/deprecated"
sALTaccount's avatar
sALTaccount committed
534
535

    def load_checkpoint(self, model_path, output_vae=True, output_clip=True):
sALTaccount's avatar
sALTaccount committed
536
537
        for search_path in folder_paths.get_folder_paths("diffusers"):
            if os.path.exists(search_path):
538
539
540
                path = os.path.join(search_path, model_path)
                if os.path.exists(path):
                    model_path = path
sALTaccount's avatar
sALTaccount committed
541
                    break
542

543
        return comfy.diffusers_load.load_diffusers(model_path, output_vae=output_vae, output_clip=output_clip, embedding_directory=folder_paths.get_folder_paths("embeddings"))
sALTaccount's avatar
sALTaccount committed
544
545


546
547
548
549
550
551
552
553
class unCLIPCheckpointLoader:
    @classmethod
    def INPUT_TYPES(s):
        return {"required": { "ckpt_name": (folder_paths.get_filename_list("checkpoints"), ),
                             }}
    RETURN_TYPES = ("MODEL", "CLIP", "VAE", "CLIP_VISION")
    FUNCTION = "load_checkpoint"

554
    CATEGORY = "loaders"
555
556
557
558
559
560

    def load_checkpoint(self, ckpt_name, output_vae=True, output_clip=True):
        ckpt_path = folder_paths.get_full_path("checkpoints", ckpt_name)
        out = comfy.sd.load_checkpoint_guess_config(ckpt_path, output_vae=True, output_clip=True, output_clipvision=True, embedding_directory=folder_paths.get_folder_paths("embeddings"))
        return out

comfyanonymous's avatar
comfyanonymous committed
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
class CLIPSetLastLayer:
    @classmethod
    def INPUT_TYPES(s):
        return {"required": { "clip": ("CLIP", ),
                              "stop_at_clip_layer": ("INT", {"default": -1, "min": -24, "max": -1, "step": 1}),
                              }}
    RETURN_TYPES = ("CLIP",)
    FUNCTION = "set_last_layer"

    CATEGORY = "conditioning"

    def set_last_layer(self, clip, stop_at_clip_layer):
        clip = clip.clone()
        clip.clip_layer(stop_at_clip_layer)
        return (clip,)

577
class LoraLoader:
578
579
580
    def __init__(self):
        self.loaded_lora = None

581
582
583
584
    @classmethod
    def INPUT_TYPES(s):
        return {"required": { "model": ("MODEL",),
                              "clip": ("CLIP", ),
585
                              "lora_name": (folder_paths.get_filename_list("loras"), ),
586
587
                              "strength_model": ("FLOAT", {"default": 1.0, "min": -20.0, "max": 20.0, "step": 0.01}),
                              "strength_clip": ("FLOAT", {"default": 1.0, "min": -20.0, "max": 20.0, "step": 0.01}),
588
589
590
591
592
593
594
                              }}
    RETURN_TYPES = ("MODEL", "CLIP")
    FUNCTION = "load_lora"

    CATEGORY = "loaders"

    def load_lora(self, model, clip, lora_name, strength_model, strength_clip):
595
596
597
        if strength_model == 0 and strength_clip == 0:
            return (model, clip)

598
        lora_path = folder_paths.get_full_path("loras", lora_name)
599
600
601
602
603
        lora = None
        if self.loaded_lora is not None:
            if self.loaded_lora[0] == lora_path:
                lora = self.loaded_lora[1]
            else:
604
605
606
                temp = self.loaded_lora
                self.loaded_lora = None
                del temp
607
608
609
610
611
612

        if lora is None:
            lora = comfy.utils.load_torch_file(lora_path, safe_load=True)
            self.loaded_lora = (lora_path, lora)

        model_lora, clip_lora = comfy.sd.load_lora_for_models(model, clip, lora, strength_model, strength_clip)
613
614
        return (model_lora, clip_lora)

615
616
617
618
619
620
621
622
623
624
625
626
627
class LoraLoaderModelOnly(LoraLoader):
    @classmethod
    def INPUT_TYPES(s):
        return {"required": { "model": ("MODEL",),
                              "lora_name": (folder_paths.get_filename_list("loras"), ),
                              "strength_model": ("FLOAT", {"default": 1.0, "min": -20.0, "max": 20.0, "step": 0.01}),
                              }}
    RETURN_TYPES = ("MODEL",)
    FUNCTION = "load_lora_model_only"

    def load_lora_model_only(self, model, lora_name, strength_model):
        return (self.load_lora(model, None, lora_name, strength_model, 0)[0],)

comfyanonymous's avatar
comfyanonymous committed
628
class VAELoader:
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
    @staticmethod
    def vae_list():
        vaes = folder_paths.get_filename_list("vae")
        approx_vaes = folder_paths.get_filename_list("vae_approx")
        sdxl_taesd_enc = False
        sdxl_taesd_dec = False
        sd1_taesd_enc = False
        sd1_taesd_dec = False

        for v in approx_vaes:
            if v.startswith("taesd_decoder."):
                sd1_taesd_dec = True
            elif v.startswith("taesd_encoder."):
                sd1_taesd_enc = True
            elif v.startswith("taesdxl_decoder."):
                sdxl_taesd_dec = True
            elif v.startswith("taesdxl_encoder."):
                sdxl_taesd_enc = True
        if sd1_taesd_dec and sd1_taesd_enc:
            vaes.append("taesd")
        if sdxl_taesd_dec and sdxl_taesd_enc:
            vaes.append("taesdxl")
        return vaes

    @staticmethod
    def load_taesd(name):
        sd = {}
        approx_vaes = folder_paths.get_filename_list("vae_approx")

        encoder = next(filter(lambda a: a.startswith("{}_encoder.".format(name)), approx_vaes))
        decoder = next(filter(lambda a: a.startswith("{}_decoder.".format(name)), approx_vaes))

        enc = comfy.utils.load_torch_file(folder_paths.get_full_path("vae_approx", encoder))
        for k in enc:
            sd["taesd_encoder.{}".format(k)] = enc[k]

        dec = comfy.utils.load_torch_file(folder_paths.get_full_path("vae_approx", decoder))
        for k in dec:
            sd["taesd_decoder.{}".format(k)] = dec[k]

        if name == "taesd":
            sd["vae_scale"] = torch.tensor(0.18215)
        elif name == "taesdxl":
            sd["vae_scale"] = torch.tensor(0.13025)
        return sd

comfyanonymous's avatar
comfyanonymous committed
675
676
    @classmethod
    def INPUT_TYPES(s):
677
        return {"required": { "vae_name": (s.vae_list(), )}}
comfyanonymous's avatar
comfyanonymous committed
678
679
680
    RETURN_TYPES = ("VAE",)
    FUNCTION = "load_vae"

681
682
    CATEGORY = "loaders"

comfyanonymous's avatar
comfyanonymous committed
683
684
    #TODO: scale factor?
    def load_vae(self, vae_name):
685
686
687
688
689
        if vae_name in ["taesd", "taesdxl"]:
            sd = self.load_taesd(vae_name)
        else:
            vae_path = folder_paths.get_full_path("vae", vae_name)
            sd = comfy.utils.load_torch_file(vae_path)
comfyanonymous's avatar
comfyanonymous committed
690
        vae = comfy.sd.VAE(sd=sd)
comfyanonymous's avatar
comfyanonymous committed
691
692
        return (vae,)

comfyanonymous's avatar
comfyanonymous committed
693
694
695
class ControlNetLoader:
    @classmethod
    def INPUT_TYPES(s):
696
        return {"required": { "control_net_name": (folder_paths.get_filename_list("controlnet"), )}}
comfyanonymous's avatar
comfyanonymous committed
697
698
699
700
701
702
703

    RETURN_TYPES = ("CONTROL_NET",)
    FUNCTION = "load_controlnet"

    CATEGORY = "loaders"

    def load_controlnet(self, control_net_name):
704
        controlnet_path = folder_paths.get_full_path("controlnet", control_net_name)
705
        controlnet = comfy.controlnet.load_controlnet(controlnet_path)
comfyanonymous's avatar
comfyanonymous committed
706
707
        return (controlnet,)

708
709
710
711
class DiffControlNetLoader:
    @classmethod
    def INPUT_TYPES(s):
        return {"required": { "model": ("MODEL",),
712
                              "control_net_name": (folder_paths.get_filename_list("controlnet"), )}}
713
714
715
716
717
718
719

    RETURN_TYPES = ("CONTROL_NET",)
    FUNCTION = "load_controlnet"

    CATEGORY = "loaders"

    def load_controlnet(self, model, control_net_name):
720
        controlnet_path = folder_paths.get_full_path("controlnet", control_net_name)
721
        controlnet = comfy.controlnet.load_controlnet(controlnet_path, model)
722
723
        return (controlnet,)

comfyanonymous's avatar
comfyanonymous committed
724
725
726
727

class ControlNetApply:
    @classmethod
    def INPUT_TYPES(s):
728
729
730
731
732
        return {"required": {"conditioning": ("CONDITIONING", ),
                             "control_net": ("CONTROL_NET", ),
                             "image": ("IMAGE", ),
                             "strength": ("FLOAT", {"default": 1.0, "min": 0.0, "max": 10.0, "step": 0.01})
                             }}
comfyanonymous's avatar
comfyanonymous committed
733
734
735
736
737
    RETURN_TYPES = ("CONDITIONING",)
    FUNCTION = "apply_controlnet"

    CATEGORY = "conditioning"

738
    def apply_controlnet(self, conditioning, control_net, image, strength):
739
740
741
        if strength == 0:
            return (conditioning, )

comfyanonymous's avatar
comfyanonymous committed
742
743
744
745
        c = []
        control_hint = image.movedim(-1,1)
        for t in conditioning:
            n = [t[0], t[1].copy()]
comfyanonymous's avatar
comfyanonymous committed
746
747
748
749
            c_net = control_net.copy().set_cond_hint(control_hint, strength)
            if 'control' in t[1]:
                c_net.set_previous_controlnet(t[1]['control'])
            n[1]['control'] = c_net
750
            n[1]['control_apply_to_uncond'] = True
comfyanonymous's avatar
comfyanonymous committed
751
752
753
            c.append(n)
        return (c, )

754
755
756
757
758
759
760
761
762

class ControlNetApplyAdvanced:
    @classmethod
    def INPUT_TYPES(s):
        return {"required": {"positive": ("CONDITIONING", ),
                             "negative": ("CONDITIONING", ),
                             "control_net": ("CONTROL_NET", ),
                             "image": ("IMAGE", ),
                             "strength": ("FLOAT", {"default": 1.0, "min": 0.0, "max": 10.0, "step": 0.01}),
763
764
                             "start_percent": ("FLOAT", {"default": 0.0, "min": 0.0, "max": 1.0, "step": 0.001}),
                             "end_percent": ("FLOAT", {"default": 1.0, "min": 0.0, "max": 1.0, "step": 0.001})
765
766
767
768
769
770
771
772
                             }}

    RETURN_TYPES = ("CONDITIONING","CONDITIONING")
    RETURN_NAMES = ("positive", "negative")
    FUNCTION = "apply_controlnet"

    CATEGORY = "conditioning"

773
    def apply_controlnet(self, positive, negative, control_net, image, strength, start_percent, end_percent):
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
        if strength == 0:
            return (positive, negative)

        control_hint = image.movedim(-1,1)
        cnets = {}

        out = []
        for conditioning in [positive, negative]:
            c = []
            for t in conditioning:
                d = t[1].copy()

                prev_cnet = d.get('control', None)
                if prev_cnet in cnets:
                    c_net = cnets[prev_cnet]
                else:
790
                    c_net = control_net.copy().set_cond_hint(control_hint, strength, (start_percent, end_percent))
791
792
793
794
795
796
797
798
799
800
801
                    c_net.set_previous_controlnet(prev_cnet)
                    cnets[prev_cnet] = c_net

                d['control'] = c_net
                d['control_apply_to_uncond'] = False
                n = [t[0], d]
                c.append(n)
            out.append(c)
        return (out[0], out[1])


802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
class UNETLoader:
    @classmethod
    def INPUT_TYPES(s):
        return {"required": { "unet_name": (folder_paths.get_filename_list("unet"), ),
                             }}
    RETURN_TYPES = ("MODEL",)
    FUNCTION = "load_unet"

    CATEGORY = "advanced/loaders"

    def load_unet(self, unet_name):
        unet_path = folder_paths.get_full_path("unet", unet_name)
        model = comfy.sd.load_unet(unet_path)
        return (model,)

817
818
819
class CLIPLoader:
    @classmethod
    def INPUT_TYPES(s):
820
        return {"required": { "clip_name": (folder_paths.get_filename_list("clip"), ),
821
                              "type": (["stable_diffusion", "stable_cascade"], ),
822
823
824
825
                             }}
    RETURN_TYPES = ("CLIP",)
    FUNCTION = "load_clip"

826
    CATEGORY = "advanced/loaders"
827

828
829
830
831
832
    def load_clip(self, clip_name, type="stable_diffusion"):
        clip_type = comfy.sd.CLIPType.STABLE_DIFFUSION
        if type == "stable_cascade":
            clip_type = comfy.sd.CLIPType.STABLE_CASCADE

833
        clip_path = folder_paths.get_full_path("clip", clip_name)
834
        clip = comfy.sd.load_clip(ckpt_paths=[clip_path], embedding_directory=folder_paths.get_folder_paths("embeddings"), clip_type=clip_type)
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
        return (clip,)

class DualCLIPLoader:
    @classmethod
    def INPUT_TYPES(s):
        return {"required": { "clip_name1": (folder_paths.get_filename_list("clip"), ), "clip_name2": (folder_paths.get_filename_list("clip"), ),
                             }}
    RETURN_TYPES = ("CLIP",)
    FUNCTION = "load_clip"

    CATEGORY = "advanced/loaders"

    def load_clip(self, clip_name1, clip_name2):
        clip_path1 = folder_paths.get_full_path("clip", clip_name1)
        clip_path2 = folder_paths.get_full_path("clip", clip_name2)
        clip = comfy.sd.load_clip(ckpt_paths=[clip_path1, clip_path2], embedding_directory=folder_paths.get_folder_paths("embeddings"))
851
852
        return (clip,)

853
854
855
class CLIPVisionLoader:
    @classmethod
    def INPUT_TYPES(s):
856
        return {"required": { "clip_name": (folder_paths.get_filename_list("clip_vision"), ),
857
858
859
860
861
862
863
                             }}
    RETURN_TYPES = ("CLIP_VISION",)
    FUNCTION = "load_clip"

    CATEGORY = "loaders"

    def load_clip(self, clip_name):
864
        clip_path = folder_paths.get_full_path("clip_vision", clip_name)
865
        clip_vision = comfy.clip_vision.load(clip_path)
866
867
868
869
870
871
872
873
        return (clip_vision,)

class CLIPVisionEncode:
    @classmethod
    def INPUT_TYPES(s):
        return {"required": { "clip_vision": ("CLIP_VISION",),
                              "image": ("IMAGE",)
                             }}
comfyanonymous's avatar
comfyanonymous committed
874
    RETURN_TYPES = ("CLIP_VISION_OUTPUT",)
875
876
    FUNCTION = "encode"

877
    CATEGORY = "conditioning"
878
879
880
881
882
883
884
885

    def encode(self, clip_vision, image):
        output = clip_vision.encode_image(image)
        return (output,)

class StyleModelLoader:
    @classmethod
    def INPUT_TYPES(s):
886
        return {"required": { "style_model_name": (folder_paths.get_filename_list("style_models"), )}}
887
888
889
890
891
892
893

    RETURN_TYPES = ("STYLE_MODEL",)
    FUNCTION = "load_style_model"

    CATEGORY = "loaders"

    def load_style_model(self, style_model_name):
894
        style_model_path = folder_paths.get_full_path("style_models", style_model_name)
895
896
897
898
899
900
901
        style_model = comfy.sd.load_style_model(style_model_path)
        return (style_model,)


class StyleModelApply:
    @classmethod
    def INPUT_TYPES(s):
902
903
904
        return {"required": {"conditioning": ("CONDITIONING", ),
                             "style_model": ("STYLE_MODEL", ),
                             "clip_vision_output": ("CLIP_VISION_OUTPUT", ),
905
906
907
908
                             }}
    RETURN_TYPES = ("CONDITIONING",)
    FUNCTION = "apply_stylemodel"

comfyanonymous's avatar
comfyanonymous committed
909
    CATEGORY = "conditioning/style_model"
910

911
    def apply_stylemodel(self, clip_vision_output, style_model, conditioning):
912
        cond = style_model.get_cond(clip_vision_output).flatten(start_dim=0, end_dim=1).unsqueeze(dim=0)
913
        c = []
914
915
        for t in conditioning:
            n = [torch.cat((t[0], cond), dim=1), t[1].copy()]
916
917
918
            c.append(n)
        return (c, )

919
920
921
922
923
924
class unCLIPConditioning:
    @classmethod
    def INPUT_TYPES(s):
        return {"required": {"conditioning": ("CONDITIONING", ),
                             "clip_vision_output": ("CLIP_VISION_OUTPUT", ),
                             "strength": ("FLOAT", {"default": 1.0, "min": -10.0, "max": 10.0, "step": 0.01}),
925
                             "noise_augmentation": ("FLOAT", {"default": 0.0, "min": 0.0, "max": 1.0, "step": 0.01}),
926
927
928
929
                             }}
    RETURN_TYPES = ("CONDITIONING",)
    FUNCTION = "apply_adm"

930
    CATEGORY = "conditioning"
931

932
    def apply_adm(self, conditioning, clip_vision_output, strength, noise_augmentation):
933
934
935
        if strength == 0:
            return (conditioning, )

936
937
938
        c = []
        for t in conditioning:
            o = t[1].copy()
939
940
941
            x = {"clip_vision_output": clip_vision_output, "strength": strength, "noise_augmentation": noise_augmentation}
            if "unclip_conditioning" in o:
                o["unclip_conditioning"] = o["unclip_conditioning"][:] + [x]
942
            else:
943
                o["unclip_conditioning"] = [x]
944
945
946
947
            n = [t[0], o]
            c.append(n)
        return (c, )

948
949
950
951
952
953
954
955
class GLIGENLoader:
    @classmethod
    def INPUT_TYPES(s):
        return {"required": { "gligen_name": (folder_paths.get_filename_list("gligen"), )}}

    RETURN_TYPES = ("GLIGEN",)
    FUNCTION = "load_gligen"

comfyanonymous's avatar
comfyanonymous committed
956
    CATEGORY = "loaders"
957
958
959
960
961
962
963
964
965
966
967
968

    def load_gligen(self, gligen_name):
        gligen_path = folder_paths.get_full_path("gligen", gligen_name)
        gligen = comfy.sd.load_gligen(gligen_path)
        return (gligen,)

class GLIGENTextBoxApply:
    @classmethod
    def INPUT_TYPES(s):
        return {"required": {"conditioning_to": ("CONDITIONING", ),
                              "clip": ("CLIP", ),
                              "gligen_textbox_model": ("GLIGEN", ),
969
                              "text": ("STRING", {"multiline": True, "dynamicPrompts": True}),
970
971
972
973
974
975
976
977
                              "width": ("INT", {"default": 64, "min": 8, "max": MAX_RESOLUTION, "step": 8}),
                              "height": ("INT", {"default": 64, "min": 8, "max": MAX_RESOLUTION, "step": 8}),
                              "x": ("INT", {"default": 0, "min": 0, "max": MAX_RESOLUTION, "step": 8}),
                              "y": ("INT", {"default": 0, "min": 0, "max": MAX_RESOLUTION, "step": 8}),
                             }}
    RETURN_TYPES = ("CONDITIONING",)
    FUNCTION = "append"

comfyanonymous's avatar
comfyanonymous committed
978
    CATEGORY = "conditioning/gligen"
979
980
981

    def append(self, conditioning_to, clip, gligen_textbox_model, text, width, height, x, y):
        c = []
982
        cond, cond_pooled = clip.encode_from_tokens(clip.tokenize(text), return_pooled="unprojected")
983
984
985
986
987
988
989
990
991
992
        for t in conditioning_to:
            n = [t[0], t[1].copy()]
            position_params = [(cond_pooled, height // 8, width // 8, y // 8, x // 8)]
            prev = []
            if "gligen" in n[1]:
                prev = n[1]['gligen'][2]

            n[1]['gligen'] = ("position", gligen_textbox_model, prev + position_params)
            c.append(n)
        return (c, )
993

comfyanonymous's avatar
comfyanonymous committed
994
class EmptyLatentImage:
995
996
    def __init__(self):
        self.device = comfy.model_management.intermediate_device()
comfyanonymous's avatar
comfyanonymous committed
997
998
999

    @classmethod
    def INPUT_TYPES(s):
1000
1001
        return {"required": { "width": ("INT", {"default": 512, "min": 16, "max": MAX_RESOLUTION, "step": 8}),
                              "height": ("INT", {"default": 512, "min": 16, "max": MAX_RESOLUTION, "step": 8}),
comfyanonymous's avatar
comfyanonymous committed
1002
                              "batch_size": ("INT", {"default": 1, "min": 1, "max": 4096})}}
comfyanonymous's avatar
comfyanonymous committed
1003
1004
1005
    RETURN_TYPES = ("LATENT",)
    FUNCTION = "generate"

1006
1007
    CATEGORY = "latent"

comfyanonymous's avatar
comfyanonymous committed
1008
    def generate(self, width, height, batch_size=1):
1009
        latent = torch.zeros([batch_size, 4, height // 8, width // 8], device=self.device)
1010
        return ({"samples":latent}, )
comfyanonymous's avatar
comfyanonymous committed
1011

comfyanonymous's avatar
comfyanonymous committed
1012

1013
1014
1015
1016
1017
class LatentFromBatch:
    @classmethod
    def INPUT_TYPES(s):
        return {"required": { "samples": ("LATENT",),
                              "batch_index": ("INT", {"default": 0, "min": 0, "max": 63}),
1018
                              "length": ("INT", {"default": 1, "min": 1, "max": 64}),
1019
1020
                              }}
    RETURN_TYPES = ("LATENT",)
1021
    FUNCTION = "frombatch"
1022

1023
    CATEGORY = "latent/batch"
1024

1025
    def frombatch(self, samples, batch_index, length):
1026
1027
1028
        s = samples.copy()
        s_in = samples["samples"]
        batch_index = min(s_in.shape[0] - 1, batch_index)
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
1066
1067
1068
        length = min(s_in.shape[0] - batch_index, length)
        s["samples"] = s_in[batch_index:batch_index + length].clone()
        if "noise_mask" in samples:
            masks = samples["noise_mask"]
            if masks.shape[0] == 1:
                s["noise_mask"] = masks.clone()
            else:
                if masks.shape[0] < s_in.shape[0]:
                    masks = masks.repeat(math.ceil(s_in.shape[0] / masks.shape[0]), 1, 1, 1)[:s_in.shape[0]]
                s["noise_mask"] = masks[batch_index:batch_index + length].clone()
        if "batch_index" not in s:
            s["batch_index"] = [x for x in range(batch_index, batch_index+length)]
        else:
            s["batch_index"] = samples["batch_index"][batch_index:batch_index + length]
        return (s,)
    
class RepeatLatentBatch:
    @classmethod
    def INPUT_TYPES(s):
        return {"required": { "samples": ("LATENT",),
                              "amount": ("INT", {"default": 1, "min": 1, "max": 64}),
                              }}
    RETURN_TYPES = ("LATENT",)
    FUNCTION = "repeat"

    CATEGORY = "latent/batch"

    def repeat(self, samples, amount):
        s = samples.copy()
        s_in = samples["samples"]
        
        s["samples"] = s_in.repeat((amount, 1,1,1))
        if "noise_mask" in samples and samples["noise_mask"].shape[0] > 1:
            masks = samples["noise_mask"]
            if masks.shape[0] < s_in.shape[0]:
                masks = masks.repeat(math.ceil(s_in.shape[0] / masks.shape[0]), 1, 1, 1)[:s_in.shape[0]]
            s["noise_mask"] = samples["noise_mask"].repeat((amount, 1,1,1))
        if "batch_index" in s:
            offset = max(s["batch_index"]) - min(s["batch_index"]) + 1
            s["batch_index"] = s["batch_index"] + [x + (i * offset) for i in range(1, amount) for x in s["batch_index"]]
1069
        return (s,)
comfyanonymous's avatar
comfyanonymous committed
1070

comfyanonymous's avatar
comfyanonymous committed
1071
class LatentUpscale:
comfyanonymous's avatar
comfyanonymous committed
1072
    upscale_methods = ["nearest-exact", "bilinear", "area", "bicubic", "bislerp"]
1073
    crop_methods = ["disabled", "center"]
comfyanonymous's avatar
comfyanonymous committed
1074
1075
1076
1077

    @classmethod
    def INPUT_TYPES(s):
        return {"required": { "samples": ("LATENT",), "upscale_method": (s.upscale_methods,),
1078
1079
                              "width": ("INT", {"default": 512, "min": 0, "max": MAX_RESOLUTION, "step": 8}),
                              "height": ("INT", {"default": 512, "min": 0, "max": MAX_RESOLUTION, "step": 8}),
1080
                              "crop": (s.crop_methods,)}}
comfyanonymous's avatar
comfyanonymous committed
1081
1082
1083
    RETURN_TYPES = ("LATENT",)
    FUNCTION = "upscale"

1084
1085
    CATEGORY = "latent"

1086
    def upscale(self, samples, upscale_method, width, height, crop):
1087
1088
1089
1090
1091
1092
1093
1094
1095
1096
1097
1098
1099
1100
1101
1102
        if width == 0 and height == 0:
            s = samples
        else:
            s = samples.copy()

            if width == 0:
                height = max(64, height)
                width = max(64, round(samples["samples"].shape[3] * height / samples["samples"].shape[2]))
            elif height == 0:
                width = max(64, width)
                height = max(64, round(samples["samples"].shape[2] * width / samples["samples"].shape[3]))
            else:
                width = max(64, width)
                height = max(64, height)

            s["samples"] = comfy.utils.common_upscale(samples["samples"], width // 8, height // 8, upscale_method, crop)
comfyanonymous's avatar
comfyanonymous committed
1103
1104
        return (s,)

comfyanonymous's avatar
comfyanonymous committed
1105
class LatentUpscaleBy:
comfyanonymous's avatar
comfyanonymous committed
1106
    upscale_methods = ["nearest-exact", "bilinear", "area", "bicubic", "bislerp"]
comfyanonymous's avatar
comfyanonymous committed
1107
1108
1109
1110
1111
1112
1113
1114
1115
1116
1117
1118
1119
1120
1121
1122
1123

    @classmethod
    def INPUT_TYPES(s):
        return {"required": { "samples": ("LATENT",), "upscale_method": (s.upscale_methods,),
                              "scale_by": ("FLOAT", {"default": 1.5, "min": 0.01, "max": 8.0, "step": 0.01}),}}
    RETURN_TYPES = ("LATENT",)
    FUNCTION = "upscale"

    CATEGORY = "latent"

    def upscale(self, samples, upscale_method, scale_by):
        s = samples.copy()
        width = round(samples["samples"].shape[3] * scale_by)
        height = round(samples["samples"].shape[2] * scale_by)
        s["samples"] = comfy.utils.common_upscale(samples["samples"], width, height, upscale_method, "disabled")
        return (s,)

comfyanonymous's avatar
comfyanonymous committed
1124
1125
1126
1127
1128
1129
1130
1131
1132
class LatentRotate:
    @classmethod
    def INPUT_TYPES(s):
        return {"required": { "samples": ("LATENT",),
                              "rotation": (["none", "90 degrees", "180 degrees", "270 degrees"],),
                              }}
    RETURN_TYPES = ("LATENT",)
    FUNCTION = "rotate"

comfyanonymous's avatar
comfyanonymous committed
1133
    CATEGORY = "latent/transform"
comfyanonymous's avatar
comfyanonymous committed
1134
1135

    def rotate(self, samples, rotation):
1136
        s = samples.copy()
comfyanonymous's avatar
comfyanonymous committed
1137
1138
1139
1140
1141
1142
1143
1144
        rotate_by = 0
        if rotation.startswith("90"):
            rotate_by = 1
        elif rotation.startswith("180"):
            rotate_by = 2
        elif rotation.startswith("270"):
            rotate_by = 3

1145
        s["samples"] = torch.rot90(samples["samples"], k=rotate_by, dims=[3, 2])
comfyanonymous's avatar
comfyanonymous committed
1146
        return (s,)
comfyanonymous's avatar
comfyanonymous committed
1147
1148
1149
1150
1151
1152
1153
1154
1155
1156

class LatentFlip:
    @classmethod
    def INPUT_TYPES(s):
        return {"required": { "samples": ("LATENT",),
                              "flip_method": (["x-axis: vertically", "y-axis: horizontally"],),
                              }}
    RETURN_TYPES = ("LATENT",)
    FUNCTION = "flip"

comfyanonymous's avatar
comfyanonymous committed
1157
    CATEGORY = "latent/transform"
comfyanonymous's avatar
comfyanonymous committed
1158
1159

    def flip(self, samples, flip_method):
1160
        s = samples.copy()
comfyanonymous's avatar
comfyanonymous committed
1161
        if flip_method.startswith("x"):
1162
            s["samples"] = torch.flip(samples["samples"], dims=[2])
comfyanonymous's avatar
comfyanonymous committed
1163
        elif flip_method.startswith("y"):
1164
            s["samples"] = torch.flip(samples["samples"], dims=[3])
comfyanonymous's avatar
comfyanonymous committed
1165
1166

        return (s,)
comfyanonymous's avatar
comfyanonymous committed
1167
1168
1169
1170

class LatentComposite:
    @classmethod
    def INPUT_TYPES(s):
comfyanonymous's avatar
comfyanonymous committed
1171
1172
1173
1174
1175
1176
        return {"required": { "samples_to": ("LATENT",),
                              "samples_from": ("LATENT",),
                              "x": ("INT", {"default": 0, "min": 0, "max": MAX_RESOLUTION, "step": 8}),
                              "y": ("INT", {"default": 0, "min": 0, "max": MAX_RESOLUTION, "step": 8}),
                              "feather": ("INT", {"default": 0, "min": 0, "max": MAX_RESOLUTION, "step": 8}),
                              }}
comfyanonymous's avatar
comfyanonymous committed
1177
1178
1179
1180
1181
    RETURN_TYPES = ("LATENT",)
    FUNCTION = "composite"

    CATEGORY = "latent"

comfyanonymous's avatar
comfyanonymous committed
1182
1183
1184
    def composite(self, samples_to, samples_from, x, y, composite_method="normal", feather=0):
        x =  x // 8
        y = y // 8
1185
        feather = feather // 8
comfyanonymous's avatar
comfyanonymous committed
1186
1187
1188
1189
1190
1191
1192
1193
1194
1195
1196
1197
1198
1199
1200
1201
1202
1203
1204
1205
1206
1207
1208
        samples_out = samples_to.copy()
        s = samples_to["samples"].clone()
        samples_to = samples_to["samples"]
        samples_from = samples_from["samples"]
        if feather == 0:
            s[:,:,y:y+samples_from.shape[2],x:x+samples_from.shape[3]] = samples_from[:,:,:samples_to.shape[2] - y, :samples_to.shape[3] - x]
        else:
            samples_from = samples_from[:,:,:samples_to.shape[2] - y, :samples_to.shape[3] - x]
            mask = torch.ones_like(samples_from)
            for t in range(feather):
                if y != 0:
                    mask[:,:,t:1+t,:] *= ((1.0/feather) * (t + 1))

                if y + samples_from.shape[2] < samples_to.shape[2]:
                    mask[:,:,mask.shape[2] -1 -t: mask.shape[2]-t,:] *= ((1.0/feather) * (t + 1))
                if x != 0:
                    mask[:,:,:,t:1+t] *= ((1.0/feather) * (t + 1))
                if x + samples_from.shape[3] < samples_to.shape[3]:
                    mask[:,:,:,mask.shape[3]- 1 - t: mask.shape[3]- t] *= ((1.0/feather) * (t + 1))
            rev_mask = torch.ones_like(mask) - mask
            s[:,:,y:y+samples_from.shape[2],x:x+samples_from.shape[3]] = samples_from[:,:,:samples_to.shape[2] - y, :samples_to.shape[3] - x] * mask + s[:,:,y:y+samples_from.shape[2],x:x+samples_from.shape[3]] * rev_mask
        samples_out["samples"] = s
        return (samples_out,)
comfyanonymous's avatar
comfyanonymous committed
1209

1210
1211
1212
1213
class LatentBlend:
    @classmethod
    def INPUT_TYPES(s):
        return {"required": {
1214
1215
            "samples1": ("LATENT",),
            "samples2": ("LATENT",),
1216
1217
1218
1219
1220
1221
1222
1223
1224
1225
1226
1227
1228
            "blend_factor": ("FLOAT", {
                "default": 0.5,
                "min": 0,
                "max": 1,
                "step": 0.01
            }),
        }}

    RETURN_TYPES = ("LATENT",)
    FUNCTION = "blend"

    CATEGORY = "_for_testing"

1229
    def blend(self, samples1, samples2, blend_factor:float, blend_mode: str="normal"):
1230

1231
1232
1233
        samples_out = samples1.copy()
        samples1 = samples1["samples"]
        samples2 = samples2["samples"]
1234

1235
1236
1237
1238
        if samples1.shape != samples2.shape:
            samples2.permute(0, 3, 1, 2)
            samples2 = comfy.utils.common_upscale(samples2, samples1.shape[3], samples1.shape[2], 'bicubic', crop='center')
            samples2.permute(0, 2, 3, 1)
1239

1240
1241
        samples_blended = self.blend_mode(samples1, samples2, blend_mode)
        samples_blended = samples1 * blend_factor + samples_blended * (1 - blend_factor)
1242
1243
1244
1245
1246
1247
1248
1249
1250
        samples_out["samples"] = samples_blended
        return (samples_out,)

    def blend_mode(self, img1, img2, mode):
        if mode == "normal":
            return img2
        else:
            raise ValueError(f"Unsupported blend mode: {mode}")

comfyanonymous's avatar
comfyanonymous committed
1251
1252
1253
1254
class LatentCrop:
    @classmethod
    def INPUT_TYPES(s):
        return {"required": { "samples": ("LATENT",),
1255
1256
                              "width": ("INT", {"default": 512, "min": 64, "max": MAX_RESOLUTION, "step": 8}),
                              "height": ("INT", {"default": 512, "min": 64, "max": MAX_RESOLUTION, "step": 8}),
1257
1258
                              "x": ("INT", {"default": 0, "min": 0, "max": MAX_RESOLUTION, "step": 8}),
                              "y": ("INT", {"default": 0, "min": 0, "max": MAX_RESOLUTION, "step": 8}),
comfyanonymous's avatar
comfyanonymous committed
1259
1260
1261
1262
                              }}
    RETURN_TYPES = ("LATENT",)
    FUNCTION = "crop"

comfyanonymous's avatar
comfyanonymous committed
1263
    CATEGORY = "latent/transform"
comfyanonymous's avatar
comfyanonymous committed
1264
1265

    def crop(self, samples, width, height, x, y):
1266
1267
        s = samples.copy()
        samples = samples['samples']
comfyanonymous's avatar
comfyanonymous committed
1268
1269
1270
1271
1272
1273
1274
1275
1276
1277
1278
1279
1280
        x =  x // 8
        y = y // 8

        #enfonce minimum size of 64
        if x > (samples.shape[3] - 8):
            x = samples.shape[3] - 8
        if y > (samples.shape[2] - 8):
            y = samples.shape[2] - 8

        new_height = height // 8
        new_width = width // 8
        to_x = new_width + x
        to_y = new_height + y
1281
        s['samples'] = samples[:,:,y:to_y, x:to_x]
comfyanonymous's avatar
comfyanonymous committed
1282
1283
        return (s,)

1284
1285
1286
1287
1288
1289
1290
1291
1292
class SetLatentNoiseMask:
    @classmethod
    def INPUT_TYPES(s):
        return {"required": { "samples": ("LATENT",),
                              "mask": ("MASK",),
                              }}
    RETURN_TYPES = ("LATENT",)
    FUNCTION = "set_mask"

1293
    CATEGORY = "latent/inpaint"
1294
1295
1296

    def set_mask(self, samples, mask):
        s = samples.copy()
1297
        s["noise_mask"] = mask.reshape((-1, 1, mask.shape[-2], mask.shape[-1]))
1298
1299
        return (s,)

space-nuko's avatar
space-nuko committed
1300
def common_ksampler(model, seed, steps, cfg, sampler_name, scheduler, positive, negative, latent, denoise=1.0, disable_noise=False, start_step=None, last_step=None, force_full_denoise=False):
1301
    latent_image = latent["samples"]
comfyanonymous's avatar
comfyanonymous committed
1302
1303
1304
    if disable_noise:
        noise = torch.zeros(latent_image.size(), dtype=latent_image.dtype, layout=latent_image.layout, device="cpu")
    else:
1305
1306
        batch_inds = latent["batch_index"] if "batch_index" in latent else None
        noise = comfy.sample.prepare_noise(latent_image, seed, batch_inds)
comfyanonymous's avatar
comfyanonymous committed
1307

1308
    noise_mask = None
1309
    if "noise_mask" in latent:
1310
        noise_mask = latent["noise_mask"]
comfyanonymous's avatar
comfyanonymous committed
1311

1312
    callback = latent_preview.prepare_callback(model, steps)
1313
    disable_pbar = not comfy.utils.PROGRESS_BAR_ENABLED
1314
1315
    samples = comfy.sample.sample(model, noise, steps, cfg, sampler_name, scheduler, positive, negative, latent_image,
                                  denoise=denoise, disable_noise=disable_noise, start_step=start_step, last_step=last_step,
comfyanonymous's avatar
comfyanonymous committed
1316
                                  force_full_denoise=force_full_denoise, noise_mask=noise_mask, callback=callback, disable_pbar=disable_pbar, seed=seed)
1317
1318
1319
    out = latent.copy()
    out["samples"] = samples
    return (out, )
comfyanonymous's avatar
comfyanonymous committed
1320

comfyanonymous's avatar
comfyanonymous committed
1321
1322
1323
class KSampler:
    @classmethod
    def INPUT_TYPES(s):
comfyanonymous's avatar
comfyanonymous committed
1324
        return {"required":
comfyanonymous's avatar
comfyanonymous committed
1325
1326
1327
                    {"model": ("MODEL",),
                    "seed": ("INT", {"default": 0, "min": 0, "max": 0xffffffffffffffff}),
                    "steps": ("INT", {"default": 20, "min": 1, "max": 10000}),
1328
                    "cfg": ("FLOAT", {"default": 8.0, "min": 0.0, "max": 100.0, "step":0.1, "round": 0.01}),
comfyanonymous's avatar
comfyanonymous committed
1329
1330
1331
1332
1333
1334
                    "sampler_name": (comfy.samplers.KSampler.SAMPLERS, ),
                    "scheduler": (comfy.samplers.KSampler.SCHEDULERS, ),
                    "positive": ("CONDITIONING", ),
                    "negative": ("CONDITIONING", ),
                    "latent_image": ("LATENT", ),
                    "denoise": ("FLOAT", {"default": 1.0, "min": 0.0, "max": 1.0, "step": 0.01}),
space-nuko's avatar
space-nuko committed
1335
1336
                     }
                }
comfyanonymous's avatar
comfyanonymous committed
1337
1338
1339
1340

    RETURN_TYPES = ("LATENT",)
    FUNCTION = "sample"

1341
1342
    CATEGORY = "sampling"

space-nuko's avatar
space-nuko committed
1343
1344
    def sample(self, model, seed, steps, cfg, sampler_name, scheduler, positive, negative, latent_image, denoise=1.0):
        return common_ksampler(model, seed, steps, cfg, sampler_name, scheduler, positive, negative, latent_image, denoise=denoise)
comfyanonymous's avatar
comfyanonymous committed
1345

comfyanonymous's avatar
comfyanonymous committed
1346
1347
1348
1349
1350
1351
1352
1353
class KSamplerAdvanced:
    @classmethod
    def INPUT_TYPES(s):
        return {"required":
                    {"model": ("MODEL",),
                    "add_noise": (["enable", "disable"], ),
                    "noise_seed": ("INT", {"default": 0, "min": 0, "max": 0xffffffffffffffff}),
                    "steps": ("INT", {"default": 20, "min": 1, "max": 10000}),
1354
                    "cfg": ("FLOAT", {"default": 8.0, "min": 0.0, "max": 100.0, "step":0.1, "round": 0.01}),
comfyanonymous's avatar
comfyanonymous committed
1355
1356
1357
1358
1359
1360
1361
1362
                    "sampler_name": (comfy.samplers.KSampler.SAMPLERS, ),
                    "scheduler": (comfy.samplers.KSampler.SCHEDULERS, ),
                    "positive": ("CONDITIONING", ),
                    "negative": ("CONDITIONING", ),
                    "latent_image": ("LATENT", ),
                    "start_at_step": ("INT", {"default": 0, "min": 0, "max": 10000}),
                    "end_at_step": ("INT", {"default": 10000, "min": 0, "max": 10000}),
                    "return_with_leftover_noise": (["disable", "enable"], ),
space-nuko's avatar
space-nuko committed
1363
1364
                     }
                }
comfyanonymous's avatar
comfyanonymous committed
1365
1366
1367
1368
1369

    RETURN_TYPES = ("LATENT",)
    FUNCTION = "sample"

    CATEGORY = "sampling"
comfyanonymous's avatar
comfyanonymous committed
1370

space-nuko's avatar
space-nuko committed
1371
    def sample(self, model, add_noise, noise_seed, steps, cfg, sampler_name, scheduler, positive, negative, latent_image, start_at_step, end_at_step, return_with_leftover_noise, denoise=1.0):
comfyanonymous's avatar
comfyanonymous committed
1372
1373
1374
1375
1376
1377
        force_full_denoise = True
        if return_with_leftover_noise == "enable":
            force_full_denoise = False
        disable_noise = False
        if add_noise == "disable":
            disable_noise = True
space-nuko's avatar
space-nuko committed
1378
        return common_ksampler(model, noise_seed, steps, cfg, sampler_name, scheduler, positive, negative, latent_image, denoise=denoise, disable_noise=disable_noise, start_step=start_at_step, last_step=end_at_step, force_full_denoise=force_full_denoise)
comfyanonymous's avatar
comfyanonymous committed
1379
1380
1381

class SaveImage:
    def __init__(self):
1382
        self.output_dir = folder_paths.get_output_directory()
m957ymj75urz's avatar
m957ymj75urz committed
1383
        self.type = "output"
1384
        self.prefix_append = ""
1385
        self.compress_level = 4
comfyanonymous's avatar
comfyanonymous committed
1386
1387
1388
1389

    @classmethod
    def INPUT_TYPES(s):
        return {"required": 
1390
                    {"images": ("IMAGE", ),
pythongosssss's avatar
tidy  
pythongosssss committed
1391
                     "filename_prefix": ("STRING", {"default": "ComfyUI"})},
pythongosssss's avatar
pythongosssss committed
1392
                "hidden": {"prompt": "PROMPT", "extra_pnginfo": "EXTRA_PNGINFO"},
comfyanonymous's avatar
comfyanonymous committed
1393
1394
1395
1396
1397
1398
1399
                }

    RETURN_TYPES = ()
    FUNCTION = "save_images"

    OUTPUT_NODE = True

1400
1401
    CATEGORY = "image"

pythongosssss's avatar
tidy  
pythongosssss committed
1402
    def save_images(self, images, filename_prefix="ComfyUI", prompt=None, extra_pnginfo=None):
1403
        filename_prefix += self.prefix_append
1404
        full_output_folder, filename, counter, subfolder, filename_prefix = folder_paths.get_save_image_path(filename_prefix, self.output_dir, images[0].shape[1], images[0].shape[0])
m957ymj75urz's avatar
m957ymj75urz committed
1405
        results = list()
1406
        for (batch_number, image) in enumerate(images):
comfyanonymous's avatar
comfyanonymous committed
1407
            i = 255. * image.cpu().numpy()
1408
            img = Image.fromarray(np.clip(i, 0, 255).astype(np.uint8))
1409
1410
1411
1412
1413
1414
1415
1416
            metadata = None
            if not args.disable_metadata:
                metadata = PngInfo()
                if prompt is not None:
                    metadata.add_text("prompt", json.dumps(prompt))
                if extra_pnginfo is not None:
                    for x in extra_pnginfo:
                        metadata.add_text(x, json.dumps(extra_pnginfo[x]))
1417

1418
1419
            filename_with_batch_num = filename.replace("%batch_num%", str(batch_number))
            file = f"{filename_with_batch_num}_{counter:05}_.png"
1420
            img.save(os.path.join(full_output_folder, file), pnginfo=metadata, compress_level=self.compress_level)
m957ymj75urz's avatar
m957ymj75urz committed
1421
1422
1423
1424
            results.append({
                "filename": file,
                "subfolder": subfolder,
                "type": self.type
Gavroche CryptoRUSH's avatar
Gavroche CryptoRUSH committed
1425
            })
1426
            counter += 1
comfyanonymous's avatar
comfyanonymous committed
1427

m957ymj75urz's avatar
m957ymj75urz committed
1428
        return { "ui": { "images": results } }
comfyanonymous's avatar
comfyanonymous committed
1429

pythongosssss's avatar
pythongosssss committed
1430
1431
class PreviewImage(SaveImage):
    def __init__(self):
1432
        self.output_dir = folder_paths.get_temp_directory()
m957ymj75urz's avatar
m957ymj75urz committed
1433
        self.type = "temp"
1434
        self.prefix_append = "_temp_" + ''.join(random.choice("abcdefghijklmnopqrstupvxyz") for x in range(5))
1435
        self.compress_level = 1
pythongosssss's avatar
pythongosssss committed
1436
1437
1438

    @classmethod
    def INPUT_TYPES(s):
1439
        return {"required":
pythongosssss's avatar
pythongosssss committed
1440
1441
1442
                    {"images": ("IMAGE", ), },
                "hidden": {"prompt": "PROMPT", "extra_pnginfo": "EXTRA_PNGINFO"},
                }
1443

1444
1445
1446
class LoadImage:
    @classmethod
    def INPUT_TYPES(s):
1447
        input_dir = folder_paths.get_input_directory()
comfyanonymous's avatar
comfyanonymous committed
1448
        files = [f for f in os.listdir(input_dir) if os.path.isfile(os.path.join(input_dir, f))]
1449
        return {"required":
1450
                    {"image": (sorted(files), {"image_upload": True})},
1451
                }
1452
1453

    CATEGORY = "image"
1454

1455
    RETURN_TYPES = ("IMAGE", "MASK")
1456
1457
    FUNCTION = "load_image"
    def load_image(self, image):
1458
        image_path = folder_paths.get_annotated_filepath(image)
1459
1460
1461
1462
1463
        img = Image.open(image_path)
        output_images = []
        output_masks = []
        for i in ImageSequence.Iterator(img):
            i = ImageOps.exif_transpose(i)
1464
1465
            if i.mode == 'I':
                i = i.point(lambda i: i * (1 / 255))
1466
1467
1468
1469
1470
1471
1472
1473
1474
1475
1476
1477
1478
1479
            image = i.convert("RGB")
            image = np.array(image).astype(np.float32) / 255.0
            image = torch.from_numpy(image)[None,]
            if 'A' in i.getbands():
                mask = np.array(i.getchannel('A')).astype(np.float32) / 255.0
                mask = 1. - torch.from_numpy(mask)
            else:
                mask = torch.zeros((64,64), dtype=torch.float32, device="cpu")
            output_images.append(image)
            output_masks.append(mask.unsqueeze(0))

        if len(output_images) > 1:
            output_image = torch.cat(output_images, dim=0)
            output_mask = torch.cat(output_masks, dim=0)
1480
        else:
1481
1482
1483
1484
            output_image = output_images[0]
            output_mask = output_masks[0]

        return (output_image, output_mask)
1485

1486
1487
    @classmethod
    def IS_CHANGED(s, image):
1488
        image_path = folder_paths.get_annotated_filepath(image)
1489
1490
1491
1492
        m = hashlib.sha256()
        with open(image_path, 'rb') as f:
            m.update(f.read())
        return m.digest().hex()
pythongosssss's avatar
pythongosssss committed
1493

1494
1495
1496
1497
1498
1499
1500
    @classmethod
    def VALIDATE_INPUTS(s, image):
        if not folder_paths.exists_annotated_filepath(image):
            return "Invalid image file: {}".format(image)

        return True

1501
class LoadImageMask:
1502
    _color_channels = ["alpha", "red", "green", "blue"]
1503
1504
    @classmethod
    def INPUT_TYPES(s):
1505
        input_dir = folder_paths.get_input_directory()
comfyanonymous's avatar
comfyanonymous committed
1506
        files = [f for f in os.listdir(input_dir) if os.path.isfile(os.path.join(input_dir, f))]
1507
        return {"required":
1508
                    {"image": (sorted(files), {"image_upload": True}),
1509
                     "channel": (s._color_channels, ), }
1510
1511
                }

1512
    CATEGORY = "mask"
1513
1514
1515
1516

    RETURN_TYPES = ("MASK",)
    FUNCTION = "load_image"
    def load_image(self, image, channel):
1517
        image_path = folder_paths.get_annotated_filepath(image)
1518
        i = Image.open(image_path)
1519
        i = ImageOps.exif_transpose(i)
1520
        if i.getbands() != ("R", "G", "B", "A"):
1521
1522
            if i.mode == 'I':
                i = i.point(lambda i: i * (1 / 255))
1523
            i = i.convert("RGBA")
1524
1525
1526
1527
1528
1529
1530
1531
1532
        mask = None
        c = channel[0].upper()
        if c in i.getbands():
            mask = np.array(i.getchannel(c)).astype(np.float32) / 255.0
            mask = torch.from_numpy(mask)
            if c == 'A':
                mask = 1. - mask
        else:
            mask = torch.zeros((64,64), dtype=torch.float32, device="cpu")
1533
        return (mask.unsqueeze(0),)
1534
1535
1536

    @classmethod
    def IS_CHANGED(s, image, channel):
1537
        image_path = folder_paths.get_annotated_filepath(image)
1538
1539
1540
1541
        m = hashlib.sha256()
        with open(image_path, 'rb') as f:
            m.update(f.read())
        return m.digest().hex()
pythongosssss's avatar
pythongosssss committed
1542

1543
    @classmethod
1544
    def VALIDATE_INPUTS(s, image):
1545
1546
1547
1548
1549
        if not folder_paths.exists_annotated_filepath(image):
            return "Invalid image file: {}".format(image)

        return True

comfyanonymous's avatar
comfyanonymous committed
1550
class ImageScale:
1551
    upscale_methods = ["nearest-exact", "bilinear", "area", "bicubic", "lanczos"]
comfyanonymous's avatar
comfyanonymous committed
1552
1553
1554
1555
1556
    crop_methods = ["disabled", "center"]

    @classmethod
    def INPUT_TYPES(s):
        return {"required": { "image": ("IMAGE",), "upscale_method": (s.upscale_methods,),
1557
1558
                              "width": ("INT", {"default": 512, "min": 0, "max": MAX_RESOLUTION, "step": 1}),
                              "height": ("INT", {"default": 512, "min": 0, "max": MAX_RESOLUTION, "step": 1}),
comfyanonymous's avatar
comfyanonymous committed
1559
1560
1561
1562
                              "crop": (s.crop_methods,)}}
    RETURN_TYPES = ("IMAGE",)
    FUNCTION = "upscale"

1563
    CATEGORY = "image/upscaling"
1564

comfyanonymous's avatar
comfyanonymous committed
1565
    def upscale(self, image, upscale_method, width, height, crop):
1566
1567
1568
1569
1570
1571
1572
1573
1574
1575
1576
1577
        if width == 0 and height == 0:
            s = image
        else:
            samples = image.movedim(-1,1)

            if width == 0:
                width = max(1, round(samples.shape[3] * height / samples.shape[2]))
            elif height == 0:
                height = max(1, round(samples.shape[2] * width / samples.shape[3]))

            s = comfy.utils.common_upscale(samples, width, height, upscale_method, crop)
            s = s.movedim(1,-1)
comfyanonymous's avatar
comfyanonymous committed
1578
        return (s,)
comfyanonymous's avatar
comfyanonymous committed
1579

comfyanonymous's avatar
comfyanonymous committed
1580
class ImageScaleBy:
1581
    upscale_methods = ["nearest-exact", "bilinear", "area", "bicubic", "lanczos"]
comfyanonymous's avatar
comfyanonymous committed
1582
1583
1584
1585
1586
1587
1588
1589
1590
1591
1592
1593
1594
1595
1596
1597
1598
1599

    @classmethod
    def INPUT_TYPES(s):
        return {"required": { "image": ("IMAGE",), "upscale_method": (s.upscale_methods,),
                              "scale_by": ("FLOAT", {"default": 1.0, "min": 0.01, "max": 8.0, "step": 0.01}),}}
    RETURN_TYPES = ("IMAGE",)
    FUNCTION = "upscale"

    CATEGORY = "image/upscaling"

    def upscale(self, image, upscale_method, scale_by):
        samples = image.movedim(-1,1)
        width = round(samples.shape[3] * scale_by)
        height = round(samples.shape[2] * scale_by)
        s = comfy.utils.common_upscale(samples, width, height, upscale_method, "disabled")
        s = s.movedim(1,-1)
        return (s,)

1600
1601
1602
1603
1604
1605
1606
1607
1608
1609
1610
1611
1612
1613
1614
class ImageInvert:

    @classmethod
    def INPUT_TYPES(s):
        return {"required": { "image": ("IMAGE",)}}

    RETURN_TYPES = ("IMAGE",)
    FUNCTION = "invert"

    CATEGORY = "image"

    def invert(self, image):
        s = 1.0 - image
        return (s,)

1615
1616
1617
1618
1619
1620
1621
1622
1623
1624
1625
1626
1627
1628
1629
1630
class ImageBatch:

    @classmethod
    def INPUT_TYPES(s):
        return {"required": { "image1": ("IMAGE",), "image2": ("IMAGE",)}}

    RETURN_TYPES = ("IMAGE",)
    FUNCTION = "batch"

    CATEGORY = "image"

    def batch(self, image1, image2):
        if image1.shape[1:] != image2.shape[1:]:
            image2 = comfy.utils.common_upscale(image2.movedim(-1,1), image1.shape[2], image1.shape[1], "bilinear", "center").movedim(1,-1)
        s = torch.cat((image1, image2), dim=0)
        return (s,)
1631

comfyanonymous's avatar
comfyanonymous committed
1632
1633
1634
1635
1636
1637
1638
1639
class EmptyImage:
    def __init__(self, device="cpu"):
        self.device = device

    @classmethod
    def INPUT_TYPES(s):
        return {"required": { "width": ("INT", {"default": 512, "min": 1, "max": MAX_RESOLUTION, "step": 1}),
                              "height": ("INT", {"default": 512, "min": 1, "max": MAX_RESOLUTION, "step": 1}),
comfyanonymous's avatar
comfyanonymous committed
1640
                              "batch_size": ("INT", {"default": 1, "min": 1, "max": 4096}),
comfyanonymous's avatar
comfyanonymous committed
1641
1642
1643
1644
1645
1646
1647
1648
1649
1650
1651
1652
1653
                              "color": ("INT", {"default": 0, "min": 0, "max": 0xFFFFFF, "step": 1, "display": "color"}),
                              }}
    RETURN_TYPES = ("IMAGE",)
    FUNCTION = "generate"

    CATEGORY = "image"

    def generate(self, width, height, batch_size=1, color=0):
        r = torch.full([batch_size, height, width, 1], ((color >> 16) & 0xFF) / 0xFF)
        g = torch.full([batch_size, height, width, 1], ((color >> 8) & 0xFF) / 0xFF)
        b = torch.full([batch_size, height, width, 1], ((color) & 0xFF) / 0xFF)
        return (torch.cat((r, g, b), dim=-1), )

Guo Y.K's avatar
Guo Y.K committed
1654
1655
1656
1657
1658
1659
1660
class ImagePadForOutpaint:

    @classmethod
    def INPUT_TYPES(s):
        return {
            "required": {
                "image": ("IMAGE",),
1661
1662
1663
1664
                "left": ("INT", {"default": 0, "min": 0, "max": MAX_RESOLUTION, "step": 8}),
                "top": ("INT", {"default": 0, "min": 0, "max": MAX_RESOLUTION, "step": 8}),
                "right": ("INT", {"default": 0, "min": 0, "max": MAX_RESOLUTION, "step": 8}),
                "bottom": ("INT", {"default": 0, "min": 0, "max": MAX_RESOLUTION, "step": 8}),
1665
                "feathering": ("INT", {"default": 40, "min": 0, "max": MAX_RESOLUTION, "step": 1}),
Guo Y.K's avatar
Guo Y.K committed
1666
1667
1668
1669
1670
1671
1672
1673
            }
        }

    RETURN_TYPES = ("IMAGE", "MASK")
    FUNCTION = "expand_image"

    CATEGORY = "image"

1674
    def expand_image(self, image, left, top, right, bottom, feathering):
Guo Y.K's avatar
Guo Y.K committed
1675
1676
        d1, d2, d3, d4 = image.size()

1677
        new_image = torch.ones(
Guo Y.K's avatar
Guo Y.K committed
1678
1679
            (d1, d2 + top + bottom, d3 + left + right, d4),
            dtype=torch.float32,
1680
1681
        ) * 0.5

Guo Y.K's avatar
Guo Y.K committed
1682
1683
1684
1685
1686
1687
        new_image[:, top:top + d2, left:left + d3, :] = image

        mask = torch.ones(
            (d2 + top + bottom, d3 + left + right),
            dtype=torch.float32,
        )
1688

1689
1690
1691
1692
1693
        t = torch.zeros(
            (d2, d3),
            dtype=torch.float32
        )

1694
        if feathering > 0 and feathering * 2 < d2 and feathering * 2 < d3:
1695
1696
1697
1698
1699
1700
1701
1702
1703
1704
1705
1706
1707
1708
1709
1710
1711
1712
1713

            for i in range(d2):
                for j in range(d3):
                    dt = i if top != 0 else d2
                    db = d2 - i if bottom != 0 else d2

                    dl = j if left != 0 else d3
                    dr = d3 - j if right != 0 else d3

                    d = min(dt, db, dl, dr)

                    if d >= feathering:
                        continue

                    v = (feathering - d) / feathering

                    t[i, j] = v * v

        mask[top:top + d2, left:left + d3] = t
1714

Guo Y.K's avatar
Guo Y.K committed
1715
1716
1717
        return (new_image, mask)


comfyanonymous's avatar
comfyanonymous committed
1718
1719
NODE_CLASS_MAPPINGS = {
    "KSampler": KSampler,
1720
    "CheckpointLoaderSimple": CheckpointLoaderSimple,
comfyanonymous's avatar
comfyanonymous committed
1721
    "CLIPTextEncode": CLIPTextEncode,
comfyanonymous's avatar
comfyanonymous committed
1722
    "CLIPSetLastLayer": CLIPSetLastLayer,
comfyanonymous's avatar
comfyanonymous committed
1723
1724
    "VAEDecode": VAEDecode,
    "VAEEncode": VAEEncode,
1725
    "VAEEncodeForInpaint": VAEEncodeForInpaint,
comfyanonymous's avatar
comfyanonymous committed
1726
1727
1728
    "VAELoader": VAELoader,
    "EmptyLatentImage": EmptyLatentImage,
    "LatentUpscale": LatentUpscale,
comfyanonymous's avatar
comfyanonymous committed
1729
    "LatentUpscaleBy": LatentUpscaleBy,
1730
    "LatentFromBatch": LatentFromBatch,
1731
    "RepeatLatentBatch": RepeatLatentBatch,
comfyanonymous's avatar
comfyanonymous committed
1732
    "SaveImage": SaveImage,
pythongosssss's avatar
pythongosssss committed
1733
    "PreviewImage": PreviewImage,
comfyanonymous's avatar
comfyanonymous committed
1734
    "LoadImage": LoadImage,
1735
    "LoadImageMask": LoadImageMask,
comfyanonymous's avatar
comfyanonymous committed
1736
    "ImageScale": ImageScale,
comfyanonymous's avatar
comfyanonymous committed
1737
    "ImageScaleBy": ImageScaleBy,
1738
    "ImageInvert": ImageInvert,
1739
    "ImageBatch": ImageBatch,
Guo Y.K's avatar
Guo Y.K committed
1740
    "ImagePadForOutpaint": ImagePadForOutpaint,
comfyanonymous's avatar
comfyanonymous committed
1741
    "EmptyImage": EmptyImage,
comfyanonymous's avatar
comfyanonymous committed
1742
    "ConditioningAverage": ConditioningAverage ,
comfyanonymous's avatar
comfyanonymous committed
1743
    "ConditioningCombine": ConditioningCombine,
1744
    "ConditioningConcat": ConditioningConcat,
comfyanonymous's avatar
comfyanonymous committed
1745
    "ConditioningSetArea": ConditioningSetArea,
1746
    "ConditioningSetAreaPercentage": ConditioningSetAreaPercentage,
1747
    "ConditioningSetAreaStrength": ConditioningSetAreaStrength,
Jacob Segal's avatar
Jacob Segal committed
1748
    "ConditioningSetMask": ConditioningSetMask,
comfyanonymous's avatar
comfyanonymous committed
1749
    "KSamplerAdvanced": KSamplerAdvanced,
1750
    "SetLatentNoiseMask": SetLatentNoiseMask,
comfyanonymous's avatar
comfyanonymous committed
1751
    "LatentComposite": LatentComposite,
1752
    "LatentBlend": LatentBlend,
comfyanonymous's avatar
comfyanonymous committed
1753
    "LatentRotate": LatentRotate,
comfyanonymous's avatar
comfyanonymous committed
1754
    "LatentFlip": LatentFlip,
comfyanonymous's avatar
comfyanonymous committed
1755
    "LatentCrop": LatentCrop,
1756
    "LoraLoader": LoraLoader,
1757
    "CLIPLoader": CLIPLoader,
1758
    "UNETLoader": UNETLoader,
1759
    "DualCLIPLoader": DualCLIPLoader,
1760
    "CLIPVisionEncode": CLIPVisionEncode,
1761
    "StyleModelApply": StyleModelApply,
1762
    "unCLIPConditioning": unCLIPConditioning,
comfyanonymous's avatar
comfyanonymous committed
1763
    "ControlNetApply": ControlNetApply,
1764
    "ControlNetApplyAdvanced": ControlNetApplyAdvanced,
comfyanonymous's avatar
comfyanonymous committed
1765
    "ControlNetLoader": ControlNetLoader,
1766
    "DiffControlNetLoader": DiffControlNetLoader,
comfyanonymous's avatar
comfyanonymous committed
1767
1768
    "StyleModelLoader": StyleModelLoader,
    "CLIPVisionLoader": CLIPVisionLoader,
1769
    "VAEDecodeTiled": VAEDecodeTiled,
comfyanonymous's avatar
comfyanonymous committed
1770
    "VAEEncodeTiled": VAEEncodeTiled,
1771
    "unCLIPCheckpointLoader": unCLIPCheckpointLoader,
1772
1773
    "GLIGENLoader": GLIGENLoader,
    "GLIGENTextBoxApply": GLIGENTextBoxApply,
1774
    "InpaintModelConditioning": InpaintModelConditioning,
1775

1776
    "CheckpointLoader": CheckpointLoader,
sALTaccount's avatar
sALTaccount committed
1777
    "DiffusersLoader": DiffusersLoader,
Dr.Lt.Data's avatar
Dr.Lt.Data committed
1778
1779

    "LoadLatent": LoadLatent,
1780
    "SaveLatent": SaveLatent,
1781
1782

    "ConditioningZeroOut": ConditioningZeroOut,
1783
    "ConditioningSetTimestepRange": ConditioningSetTimestepRange,
1784
    "LoraLoaderModelOnly": LoraLoaderModelOnly,
comfyanonymous's avatar
comfyanonymous committed
1785
1786
}

City's avatar
City committed
1787
1788
1789
1790
1791
NODE_DISPLAY_NAME_MAPPINGS = {
    # Sampling
    "KSampler": "KSampler",
    "KSamplerAdvanced": "KSampler (Advanced)",
    # Loaders
1792
    "CheckpointLoader": "Load Checkpoint With Config (DEPRECATED)",
comfyanonymous's avatar
Rename.  
comfyanonymous committed
1793
    "CheckpointLoaderSimple": "Load Checkpoint",
City's avatar
City committed
1794
1795
1796
1797
1798
1799
1800
1801
1802
1803
1804
1805
1806
1807
    "VAELoader": "Load VAE",
    "LoraLoader": "Load LoRA",
    "CLIPLoader": "Load CLIP",
    "ControlNetLoader": "Load ControlNet Model",
    "DiffControlNetLoader": "Load ControlNet Model (diff)",
    "StyleModelLoader": "Load Style Model",
    "CLIPVisionLoader": "Load CLIP Vision",
    "UpscaleModelLoader": "Load Upscale Model",
    # Conditioning
    "CLIPVisionEncode": "CLIP Vision Encode",
    "StyleModelApply": "Apply Style Model",
    "CLIPTextEncode": "CLIP Text Encode (Prompt)",
    "CLIPSetLastLayer": "CLIP Set Last Layer",
    "ConditioningCombine": "Conditioning (Combine)",
FizzleDorf's avatar
FizzleDorf committed
1808
    "ConditioningAverage ": "Conditioning (Average)",
1809
    "ConditioningConcat": "Conditioning (Concat)",
City's avatar
City committed
1810
    "ConditioningSetArea": "Conditioning (Set Area)",
1811
    "ConditioningSetAreaPercentage": "Conditioning (Set Area with Percentage)",
Jacob Segal's avatar
Jacob Segal committed
1812
    "ConditioningSetMask": "Conditioning (Set Mask)",
City's avatar
City committed
1813
    "ControlNetApply": "Apply ControlNet",
1814
    "ControlNetApplyAdvanced": "Apply ControlNet (Advanced)",
City's avatar
City committed
1815
1816
1817
1818
1819
1820
1821
1822
1823
1824
    # Latent
    "VAEEncodeForInpaint": "VAE Encode (for Inpainting)",
    "SetLatentNoiseMask": "Set Latent Noise Mask",
    "VAEDecode": "VAE Decode",
    "VAEEncode": "VAE Encode",
    "LatentRotate": "Rotate Latent",
    "LatentFlip": "Flip Latent",
    "LatentCrop": "Crop Latent",
    "EmptyLatentImage": "Empty Latent Image",
    "LatentUpscale": "Upscale Latent",
comfyanonymous's avatar
comfyanonymous committed
1825
    "LatentUpscaleBy": "Upscale Latent By",
City's avatar
City committed
1826
    "LatentComposite": "Latent Composite",
1827
    "LatentBlend": "Latent Blend",
1828
1829
    "LatentFromBatch" : "Latent From Batch",
    "RepeatLatentBatch": "Repeat Latent Batch",
City's avatar
City committed
1830
1831
1832
1833
1834
1835
    # Image
    "SaveImage": "Save Image",
    "PreviewImage": "Preview Image",
    "LoadImage": "Load Image",
    "LoadImageMask": "Load Image (as Mask)",
    "ImageScale": "Upscale Image",
comfyanonymous's avatar
comfyanonymous committed
1836
    "ImageScaleBy": "Upscale Image By",
City's avatar
City committed
1837
1838
1839
    "ImageUpscaleWithModel": "Upscale Image (using Model)",
    "ImageInvert": "Invert Image",
    "ImagePadForOutpaint": "Pad Image for Outpainting",
1840
    "ImageBatch": "Batch Images",
City's avatar
City committed
1841
1842
1843
1844
1845
    # _for_testing
    "VAEDecodeTiled": "VAE Decode (Tiled)",
    "VAEEncodeTiled": "VAE Encode (Tiled)",
}

1846
1847
EXTENSION_WEB_DIRS = {}

1848
def load_custom_node(module_path, ignore=set()):
1849
1850
1851
1852
1853
    module_name = os.path.basename(module_path)
    if os.path.isfile(module_path):
        sp = os.path.splitext(module_path)
        module_name = sp[0]
    try:
1854
        logging.debug("Trying to load custom node {}".format(module_path))
1855
1856
        if os.path.isfile(module_path):
            module_spec = importlib.util.spec_from_file_location(module_name, module_path)
1857
            module_dir = os.path.split(module_path)[0]
1858
1859
        else:
            module_spec = importlib.util.spec_from_file_location(module_name, os.path.join(module_path, "__init__.py"))
1860
1861
            module_dir = module_path

1862
1863
1864
        module = importlib.util.module_from_spec(module_spec)
        sys.modules[module_name] = module
        module_spec.loader.exec_module(module)
1865
1866
1867
1868
1869
1870

        if hasattr(module, "WEB_DIRECTORY") and getattr(module, "WEB_DIRECTORY") is not None:
            web_dir = os.path.abspath(os.path.join(module_dir, getattr(module, "WEB_DIRECTORY")))
            if os.path.isdir(web_dir):
                EXTENSION_WEB_DIRS[module_name] = web_dir

1871
        if hasattr(module, "NODE_CLASS_MAPPINGS") and getattr(module, "NODE_CLASS_MAPPINGS") is not None:
1872
1873
1874
            for name in module.NODE_CLASS_MAPPINGS:
                if name not in ignore:
                    NODE_CLASS_MAPPINGS[name] = module.NODE_CLASS_MAPPINGS[name]
1875
1876
            if hasattr(module, "NODE_DISPLAY_NAME_MAPPINGS") and getattr(module, "NODE_DISPLAY_NAME_MAPPINGS") is not None:
                NODE_DISPLAY_NAME_MAPPINGS.update(module.NODE_DISPLAY_NAME_MAPPINGS)
1877
            return True
1878
        else:
1879
            logging.warning(f"Skip {module_path} module for custom nodes due to the lack of NODE_CLASS_MAPPINGS.")
1880
            return False
1881
    except Exception as e:
1882
        logging.warning(traceback.format_exc())
1883
        logging.warning(f"Cannot import {module_path} module for custom nodes: {e}")
1884
        return False
1885

Hacker 17082006's avatar
Hacker 17082006 committed
1886
def load_custom_nodes():
1887
    base_node_names = set(NODE_CLASS_MAPPINGS.keys())
1888
    node_paths = folder_paths.get_folder_paths("custom_nodes")
1889
    node_import_times = []
1890
    for custom_node_path in node_paths:
Enrico Fasoli's avatar
Enrico Fasoli committed
1891
        possible_modules = os.listdir(os.path.realpath(custom_node_path))
1892
1893
1894
1895
1896
1897
        if "__pycache__" in possible_modules:
            possible_modules.remove("__pycache__")

        for possible_module in possible_modules:
            module_path = os.path.join(custom_node_path, possible_module)
            if os.path.isfile(module_path) and os.path.splitext(module_path)[1] != ".py": continue
1898
            if module_path.endswith(".disabled"): continue
1899
            time_before = time.perf_counter()
1900
            success = load_custom_node(module_path, base_node_names)
1901
            node_import_times.append((time.perf_counter() - time_before, module_path, success))
1902

1903
    if len(node_import_times) > 0:
comfyanonymous's avatar
comfyanonymous committed
1904
        logging.info("\nImport times for custom nodes:")
1905
        for n in sorted(node_import_times):
1906
1907
1908
1909
            if n[2]:
                import_message = ""
            else:
                import_message = " (IMPORT FAILED)"
comfyanonymous's avatar
comfyanonymous committed
1910
1911
            logging.info("{:6.1f} seconds{}: {}".format(n[0], import_message, n[1]))
        logging.info("")
1912

1913
def init_custom_nodes():
1914
1915
1916
1917
1918
1919
1920
    extras_dir = os.path.join(os.path.dirname(os.path.realpath(__file__)), "comfy_extras")
    extras_files = [
        "nodes_latent.py",
        "nodes_hypernetwork.py",
        "nodes_upscale_model.py",
        "nodes_post_processing.py",
        "nodes_mask.py",
1921
        "nodes_compositing.py",
1922
1923
1924
1925
1926
1927
        "nodes_rebatch.py",
        "nodes_model_merging.py",
        "nodes_tomesd.py",
        "nodes_clip_sdxl.py",
        "nodes_canny.py",
        "nodes_freelunch.py",
1928
1929
        "nodes_custom_sampler.py",
        "nodes_hypertile.py",
1930
        "nodes_model_advanced.py",
1931
        "nodes_model_downscale.py",
comfyanonymous's avatar
comfyanonymous committed
1932
        "nodes_images.py",
1933
        "nodes_video_model.py",
1934
        "nodes_sag.py",
Hari's avatar
Hari committed
1935
        "nodes_perpneg.py",
1936
        "nodes_stable3d.py",
1937
        "nodes_sdupscale.py",
1938
        "nodes_photomaker.py",
1939
        "nodes_cond.py",
1940
        "nodes_morphology.py",
comfyanonymous's avatar
comfyanonymous committed
1941
        "nodes_stable_cascade.py",
1942
        "nodes_differential_diffusion.py",
1943
        "nodes_ip2p.py",
1944
        "nodes_model_merging_model_specific.py",
1945
1946
    ]

1947
    import_failed = []
1948
    for node_file in extras_files:
1949
1950
        if not load_custom_node(os.path.join(extras_dir, node_file)):
            import_failed.append(node_file)
1951

1952
    load_custom_nodes()
1953
1954

    if len(import_failed) > 0:
1955
        logging.warning("WARNING: some comfy_extras/ nodes did not import correctly. This may be because they are missing some dependencies.\n")
1956
        for node in import_failed:
1957
1958
            logging.warning("IMPORT FAILED: {}".format(node))
        logging.warning("\nThis issue might be caused by new missing dependencies added the last time you updated ComfyUI.")
1959
        if args.windows_standalone_build:
1960
            logging.warning("Please run the update script: update/update_comfyui.bat")
1961
        else:
1962
1963
            logging.warning("Please do a: pip install -r requirements.txt")
        logging.warning("")