nodes.py 47.1 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
comfyanonymous's avatar
comfyanonymous committed
9
10
11
12
13

from PIL import Image
from PIL.PngImagePlugin import PngInfo
import numpy as np

sALTaccount's avatar
sALTaccount committed
14

comfyanonymous's avatar
comfyanonymous committed
15
sys.path.insert(0, os.path.join(os.path.dirname(os.path.realpath(__file__)), "comfy"))
comfyanonymous's avatar
comfyanonymous committed
16
17


comfyanonymous's avatar
comfyanonymous committed
18
import comfy.diffusers_convert
comfyanonymous's avatar
comfyanonymous committed
19
import comfy.samplers
20
import comfy.sample
comfyanonymous's avatar
comfyanonymous committed
21
import comfy.sd
comfyanonymous's avatar
comfyanonymous committed
22
23
import comfy.utils

24
import comfy.clip_vision
25

26
import comfy.model_management
27
import importlib
comfyanonymous's avatar
comfyanonymous committed
28

29
import folder_paths
30
31

def before_node_execution():
32
    comfy.model_management.throw_exception_if_processing_interrupted()
33

34
def interrupt_processing(value=True):
35
    comfy.model_management.interrupt_current_processing(value)
36

37
38
MAX_RESOLUTION=8192

comfyanonymous's avatar
comfyanonymous committed
39
40
41
class CLIPTextEncode:
    @classmethod
    def INPUT_TYPES(s):
42
        return {"required": {"text": ("STRING", {"multiline": True}), "clip": ("CLIP", )}}
comfyanonymous's avatar
comfyanonymous committed
43
44
45
    RETURN_TYPES = ("CONDITIONING",)
    FUNCTION = "encode"

46
47
    CATEGORY = "conditioning"

comfyanonymous's avatar
comfyanonymous committed
48
    def encode(self, clip, text):
comfyanonymous's avatar
comfyanonymous committed
49
50
51
52
53
54
55
56
57
        return ([[clip.encode(text), {}]], )

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

58
59
    CATEGORY = "conditioning"

comfyanonymous's avatar
comfyanonymous committed
60
61
62
    def combine(self, conditioning_1, conditioning_2):
        return (conditioning_1 + conditioning_2, )

FizzleDorf's avatar
FizzleDorf committed
63
64
65
class ConditioningAverage :
    @classmethod
    def INPUT_TYPES(s):
comfyanonymous's avatar
comfyanonymous committed
66
67
        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
68
69
70
71
72
73
                             }}
    RETURN_TYPES = ("CONDITIONING",)
    FUNCTION = "addWeighted"

    CATEGORY = "conditioning"

comfyanonymous's avatar
comfyanonymous committed
74
    def addWeighted(self, conditioning_to, conditioning_from, conditioning_to_strength):
FizzleDorf's avatar
FizzleDorf committed
75
        out = []
comfyanonymous's avatar
comfyanonymous committed
76
77
78
79
80
81
82
83
84
85
86
87
88
89

        if len(conditioning_from) > 1:
            print("Warning: ConditioningAverage conditioning_from contains more than 1 cond, only the first one will actually be applied to conditioning_to.")

        cond_from = conditioning_from[0][0]

        for i in range(len(conditioning_to)):
            t1 = conditioning_to[i][0]
            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))
            n = [tw, conditioning_to[i][1].copy()]
FizzleDorf's avatar
FizzleDorf committed
90
91
92
            out.append(n)
        return (out, )

comfyanonymous's avatar
comfyanonymous committed
93
94
95
96
class ConditioningSetArea:
    @classmethod
    def INPUT_TYPES(s):
        return {"required": {"conditioning": ("CONDITIONING", ),
97
98
99
100
                              "width": ("INT", {"default": 64, "min": 64, "max": MAX_RESOLUTION, "step": 64}),
                              "height": ("INT", {"default": 64, "min": 64, "max": MAX_RESOLUTION, "step": 64}),
                              "x": ("INT", {"default": 0, "min": 0, "max": MAX_RESOLUTION, "step": 64}),
                              "y": ("INT", {"default": 0, "min": 0, "max": MAX_RESOLUTION, "step": 64}),
comfyanonymous's avatar
comfyanonymous committed
101
102
103
104
105
                              "strength": ("FLOAT", {"default": 1.0, "min": 0.0, "max": 10.0, "step": 0.01}),
                             }}
    RETURN_TYPES = ("CONDITIONING",)
    FUNCTION = "append"

106
107
    CATEGORY = "conditioning"

comfyanonymous's avatar
comfyanonymous committed
108
    def append(self, conditioning, width, height, x, y, strength, min_sigma=0.0, max_sigma=99.0):
comfyanonymous's avatar
comfyanonymous committed
109
110
111
112
113
        c = []
        for t in conditioning:
            n = [t[0], t[1].copy()]
            n[1]['area'] = (height // 8, width // 8, y // 8, x // 8)
            n[1]['strength'] = strength
114
            n[1]['set_area_to_bounds'] = False
comfyanonymous's avatar
comfyanonymous committed
115
116
117
            n[1]['min_sigma'] = min_sigma
            n[1]['max_sigma'] = max_sigma
            c.append(n)
comfyanonymous's avatar
comfyanonymous committed
118
        return (c, )
comfyanonymous's avatar
comfyanonymous committed
119

Jacob Segal's avatar
Jacob Segal committed
120
121
122
123
124
125
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}),
126
                              "set_cond_area": (["default", "mask bounds"],),
Jacob Segal's avatar
Jacob Segal committed
127
128
129
130
131
132
                             }}
    RETURN_TYPES = ("CONDITIONING",)
    FUNCTION = "append"

    CATEGORY = "conditioning"

133
    def append(self, conditioning, mask, set_cond_area, strength):
Jacob Segal's avatar
Jacob Segal committed
134
        c = []
135
136
137
        set_area_to_bounds = False
        if set_cond_area != "default":
            set_area_to_bounds = True
Jacob Segal's avatar
Jacob Segal committed
138
139
140
141
142
143
        if len(mask.shape) < 3:
            mask = mask.unsqueeze(0)
        for t in conditioning:
            n = [t[0], t[1].copy()]
            _, h, w = mask.shape
            n[1]['mask'] = mask
Jacob Segal's avatar
Jacob Segal committed
144
            n[1]['set_area_to_bounds'] = set_area_to_bounds
145
            n[1]['mask_strength'] = strength
Jacob Segal's avatar
Jacob Segal committed
146
147
148
            c.append(n)
        return (c, )

comfyanonymous's avatar
comfyanonymous committed
149
150
151
152
153
154
155
156
157
158
class VAEDecode:
    def __init__(self, device="cpu"):
        self.device = device

    @classmethod
    def INPUT_TYPES(s):
        return {"required": { "samples": ("LATENT", ), "vae": ("VAE", )}}
    RETURN_TYPES = ("IMAGE",)
    FUNCTION = "decode"

159
160
    CATEGORY = "latent"

comfyanonymous's avatar
comfyanonymous committed
161
    def decode(self, vae, samples):
162
        return (vae.decode(samples["samples"]), )
comfyanonymous's avatar
comfyanonymous committed
163

164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
class VAEDecodeTiled:
    def __init__(self, device="cpu"):
        self.device = device

    @classmethod
    def INPUT_TYPES(s):
        return {"required": { "samples": ("LATENT", ), "vae": ("VAE", )}}
    RETURN_TYPES = ("IMAGE",)
    FUNCTION = "decode"

    CATEGORY = "_for_testing"

    def decode(self, vae, samples):
        return (vae.decode_tiled(samples["samples"]), )

comfyanonymous's avatar
comfyanonymous committed
179
180
181
182
183
184
185
186
187
188
class VAEEncode:
    def __init__(self, device="cpu"):
        self.device = device

    @classmethod
    def INPUT_TYPES(s):
        return {"required": { "pixels": ("IMAGE", ), "vae": ("VAE", )}}
    RETURN_TYPES = ("LATENT",)
    FUNCTION = "encode"

189
190
    CATEGORY = "latent"

comfyanonymous's avatar
comfyanonymous committed
191
    def encode(self, vae, pixels):
192
193
194
195
        x = (pixels.shape[1] // 64) * 64
        y = (pixels.shape[2] // 64) * 64
        if pixels.shape[1] != x or pixels.shape[2] != y:
            pixels = pixels[:,:x,:y,:]
196
197
198
        t = vae.encode(pixels[:,:,:,:3])

        return ({"samples":t}, )
comfyanonymous's avatar
comfyanonymous committed
199

comfyanonymous's avatar
comfyanonymous committed
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220

class VAEEncodeTiled:
    def __init__(self, device="cpu"):
        self.device = device

    @classmethod
    def INPUT_TYPES(s):
        return {"required": { "pixels": ("IMAGE", ), "vae": ("VAE", )}}
    RETURN_TYPES = ("LATENT",)
    FUNCTION = "encode"

    CATEGORY = "_for_testing"

    def encode(self, vae, pixels):
        x = (pixels.shape[1] // 64) * 64
        y = (pixels.shape[2] // 64) * 64
        if pixels.shape[1] != x or pixels.shape[2] != y:
            pixels = pixels[:,:x,:y,:]
        t = vae.encode_tiled(pixels[:,:,:,:3])

        return ({"samples":t}, )
221
222
223
224
225
226
class VAEEncodeForInpaint:
    def __init__(self, device="cpu"):
        self.device = device

    @classmethod
    def INPUT_TYPES(s):
227
        return {"required": { "pixels": ("IMAGE", ), "vae": ("VAE", ), "mask": ("MASK", ), "grow_mask_by": ("INT", {"default": 6, "min": 0, "max": 64, "step": 1}),}}
228
229
230
231
232
    RETURN_TYPES = ("LATENT",)
    FUNCTION = "encode"

    CATEGORY = "latent/inpaint"

233
    def encode(self, vae, pixels, mask, grow_mask_by=6):
234
235
        x = (pixels.shape[1] // 64) * 64
        y = (pixels.shape[2] // 64) * 64
236
        mask = torch.nn.functional.interpolate(mask.reshape((-1, 1, mask.shape[-2], mask.shape[-1])), size=(pixels.shape[1], pixels.shape[2]), mode="bilinear")
237

238
        pixels = pixels.clone()
239
240
        if pixels.shape[1] != x or pixels.shape[2] != y:
            pixels = pixels[:,:x,:y,:]
241
            mask = mask[:,:,:x,:y]
242

243
        #grow mask by a few pixels to keep things seamless in latent space
244
245
246
247
248
249
250
251
        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)

252
        m = (1.0 - mask.round()).squeeze(1)
253
254
        for i in range(3):
            pixels[:,:,:,i] -= 0.5
255
            pixels[:,:,:,i] *= m
256
257
258
            pixels[:,:,:,i] += 0.5
        t = vae.encode(pixels)

259
        return ({"samples":t, "noise_mask": (mask_erosion[:,:,:x,:y].round())}, )
comfyanonymous's avatar
comfyanonymous committed
260
261
262
263

class CheckpointLoader:
    @classmethod
    def INPUT_TYPES(s):
264
265
        return {"required": { "config_name": (folder_paths.get_filename_list("configs"), ),
                              "ckpt_name": (folder_paths.get_filename_list("checkpoints"), )}}
comfyanonymous's avatar
comfyanonymous committed
266
267
268
    RETURN_TYPES = ("MODEL", "CLIP", "VAE")
    FUNCTION = "load_checkpoint"

269
    CATEGORY = "advanced/loaders"
270

comfyanonymous's avatar
comfyanonymous committed
271
    def load_checkpoint(self, config_name, ckpt_name, output_vae=True, output_clip=True):
272
273
        config_path = folder_paths.get_full_path("configs", config_name)
        ckpt_path = folder_paths.get_full_path("checkpoints", ckpt_name)
274
        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
275

276
277
278
class CheckpointLoaderSimple:
    @classmethod
    def INPUT_TYPES(s):
279
        return {"required": { "ckpt_name": (folder_paths.get_filename_list("checkpoints"), ),
280
281
282
283
                             }}
    RETURN_TYPES = ("MODEL", "CLIP", "VAE")
    FUNCTION = "load_checkpoint"

284
    CATEGORY = "loaders"
285

286
    def load_checkpoint(self, ckpt_name, output_vae=True, output_clip=True):
287
        ckpt_path = folder_paths.get_full_path("checkpoints", ckpt_name)
288
        out = comfy.sd.load_checkpoint_guess_config(ckpt_path, output_vae=True, output_clip=True, embedding_directory=folder_paths.get_folder_paths("embeddings"))
289
290
        return out

sALTaccount's avatar
sALTaccount committed
291
292
293
class DiffusersLoader:
    @classmethod
    def INPUT_TYPES(cls):
294
        paths = []
sALTaccount's avatar
sALTaccount committed
295
        for search_path in folder_paths.get_folder_paths("diffusers"):
296
            if os.path.exists(search_path):
sALTaccount's avatar
sALTaccount committed
297
                paths += next(os.walk(search_path))[1]
298
        return {"required": {"model_path": (paths,), }}
sALTaccount's avatar
sALTaccount committed
299
300
301
    RETURN_TYPES = ("MODEL", "CLIP", "VAE")
    FUNCTION = "load_checkpoint"

302
    CATEGORY = "advanced/loaders"
sALTaccount's avatar
sALTaccount committed
303
304

    def load_checkpoint(self, model_path, output_vae=True, output_clip=True):
sALTaccount's avatar
sALTaccount committed
305
306
307
308
309
310
        for search_path in folder_paths.get_folder_paths("diffusers"):
            if os.path.exists(search_path):
                paths = next(os.walk(search_path))[1]
                if model_path in paths:
                    model_path = os.path.join(search_path, model_path)
                    break
311

312
        return comfy.diffusers_convert.load_diffusers(model_path, fp16=comfy.model_management.should_use_fp16(), output_vae=output_vae, output_clip=output_clip, embedding_directory=folder_paths.get_folder_paths("embeddings"))
sALTaccount's avatar
sALTaccount committed
313
314


315
316
317
318
319
320
321
322
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"

323
    CATEGORY = "loaders"
324
325
326
327
328
329

    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
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
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,)

346
347
348
349
350
class LoraLoader:
    @classmethod
    def INPUT_TYPES(s):
        return {"required": { "model": ("MODEL",),
                              "clip": ("CLIP", ),
351
                              "lora_name": (folder_paths.get_filename_list("loras"), ),
352
353
                              "strength_model": ("FLOAT", {"default": 1.0, "min": -10.0, "max": 10.0, "step": 0.01}),
                              "strength_clip": ("FLOAT", {"default": 1.0, "min": -10.0, "max": 10.0, "step": 0.01}),
354
355
356
357
358
359
360
                              }}
    RETURN_TYPES = ("MODEL", "CLIP")
    FUNCTION = "load_lora"

    CATEGORY = "loaders"

    def load_lora(self, model, clip, lora_name, strength_model, strength_clip):
361
        lora_path = folder_paths.get_full_path("loras", lora_name)
362
363
364
        model_lora, clip_lora = comfy.sd.load_lora_for_models(model, clip, lora_path, strength_model, strength_clip)
        return (model_lora, clip_lora)

365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
class TomePatchModel:
    @classmethod
    def INPUT_TYPES(s):
        return {"required": { "model": ("MODEL",),
                              "ratio": ("FLOAT", {"default": 0.3, "min": 0.0, "max": 1.0, "step": 0.01}),
                              }}
    RETURN_TYPES = ("MODEL",)
    FUNCTION = "patch"

    CATEGORY = "_for_testing"

    def patch(self, model, ratio):
        m = model.clone()
        m.set_model_tomesd(ratio)
        return (m, )

comfyanonymous's avatar
comfyanonymous committed
381
382
383
class VAELoader:
    @classmethod
    def INPUT_TYPES(s):
384
        return {"required": { "vae_name": (folder_paths.get_filename_list("vae"), )}}
comfyanonymous's avatar
comfyanonymous committed
385
386
387
    RETURN_TYPES = ("VAE",)
    FUNCTION = "load_vae"

388
389
    CATEGORY = "loaders"

comfyanonymous's avatar
comfyanonymous committed
390
391
    #TODO: scale factor?
    def load_vae(self, vae_name):
392
        vae_path = folder_paths.get_full_path("vae", vae_name)
comfyanonymous's avatar
comfyanonymous committed
393
394
395
        vae = comfy.sd.VAE(ckpt_path=vae_path)
        return (vae,)

comfyanonymous's avatar
comfyanonymous committed
396
397
398
class ControlNetLoader:
    @classmethod
    def INPUT_TYPES(s):
399
        return {"required": { "control_net_name": (folder_paths.get_filename_list("controlnet"), )}}
comfyanonymous's avatar
comfyanonymous committed
400
401
402
403
404
405
406

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

    CATEGORY = "loaders"

    def load_controlnet(self, control_net_name):
407
        controlnet_path = folder_paths.get_full_path("controlnet", control_net_name)
comfyanonymous's avatar
comfyanonymous committed
408
409
410
        controlnet = comfy.sd.load_controlnet(controlnet_path)
        return (controlnet,)

411
412
413
414
class DiffControlNetLoader:
    @classmethod
    def INPUT_TYPES(s):
        return {"required": { "model": ("MODEL",),
415
                              "control_net_name": (folder_paths.get_filename_list("controlnet"), )}}
416
417
418
419
420
421
422

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

    CATEGORY = "loaders"

    def load_controlnet(self, model, control_net_name):
423
        controlnet_path = folder_paths.get_full_path("controlnet", control_net_name)
424
425
426
        controlnet = comfy.sd.load_controlnet(controlnet_path, model)
        return (controlnet,)

comfyanonymous's avatar
comfyanonymous committed
427
428
429
430

class ControlNetApply:
    @classmethod
    def INPUT_TYPES(s):
431
432
433
434
435
        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
436
437
438
439
440
    RETURN_TYPES = ("CONDITIONING",)
    FUNCTION = "apply_controlnet"

    CATEGORY = "conditioning"

441
    def apply_controlnet(self, conditioning, control_net, image, strength):
comfyanonymous's avatar
comfyanonymous committed
442
443
444
445
446
        c = []
        control_hint = image.movedim(-1,1)
        print(control_hint.shape)
        for t in conditioning:
            n = [t[0], t[1].copy()]
comfyanonymous's avatar
comfyanonymous committed
447
448
449
450
            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
comfyanonymous's avatar
comfyanonymous committed
451
452
453
            c.append(n)
        return (c, )

454
455
456
class CLIPLoader:
    @classmethod
    def INPUT_TYPES(s):
457
        return {"required": { "clip_name": (folder_paths.get_filename_list("clip"), ),
458
459
460
461
462
463
                             }}
    RETURN_TYPES = ("CLIP",)
    FUNCTION = "load_clip"

    CATEGORY = "loaders"

464
    def load_clip(self, clip_name):
465
        clip_path = folder_paths.get_full_path("clip", clip_name)
comfyanonymous's avatar
comfyanonymous committed
466
        clip = comfy.sd.load_clip(ckpt_path=clip_path, embedding_directory=folder_paths.get_folder_paths("embeddings"))
467
468
        return (clip,)

469
470
471
class CLIPVisionLoader:
    @classmethod
    def INPUT_TYPES(s):
472
        return {"required": { "clip_name": (folder_paths.get_filename_list("clip_vision"), ),
473
474
475
476
477
478
479
                             }}
    RETURN_TYPES = ("CLIP_VISION",)
    FUNCTION = "load_clip"

    CATEGORY = "loaders"

    def load_clip(self, clip_name):
480
        clip_path = folder_paths.get_full_path("clip_vision", clip_name)
481
        clip_vision = comfy.clip_vision.load(clip_path)
482
483
484
485
486
487
488
489
        return (clip_vision,)

class CLIPVisionEncode:
    @classmethod
    def INPUT_TYPES(s):
        return {"required": { "clip_vision": ("CLIP_VISION",),
                              "image": ("IMAGE",)
                             }}
comfyanonymous's avatar
comfyanonymous committed
490
    RETURN_TYPES = ("CLIP_VISION_OUTPUT",)
491
492
    FUNCTION = "encode"

493
    CATEGORY = "conditioning"
494
495
496
497
498
499
500
501

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

class StyleModelLoader:
    @classmethod
    def INPUT_TYPES(s):
502
        return {"required": { "style_model_name": (folder_paths.get_filename_list("style_models"), )}}
503
504
505
506
507
508
509

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

    CATEGORY = "loaders"

    def load_style_model(self, style_model_name):
510
        style_model_path = folder_paths.get_full_path("style_models", style_model_name)
511
512
513
514
515
516
517
        style_model = comfy.sd.load_style_model(style_model_path)
        return (style_model,)


class StyleModelApply:
    @classmethod
    def INPUT_TYPES(s):
518
519
520
        return {"required": {"conditioning": ("CONDITIONING", ),
                             "style_model": ("STYLE_MODEL", ),
                             "clip_vision_output": ("CLIP_VISION_OUTPUT", ),
521
522
523
524
                             }}
    RETURN_TYPES = ("CONDITIONING",)
    FUNCTION = "apply_stylemodel"

comfyanonymous's avatar
comfyanonymous committed
525
    CATEGORY = "conditioning/style_model"
526

527
528
    def apply_stylemodel(self, clip_vision_output, style_model, conditioning):
        cond = style_model.get_cond(clip_vision_output)
529
        c = []
530
531
        for t in conditioning:
            n = [torch.cat((t[0], cond), dim=1), t[1].copy()]
532
533
534
            c.append(n)
        return (c, )

535
536
537
538
539
540
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}),
541
                             "noise_augmentation": ("FLOAT", {"default": 0.0, "min": 0.0, "max": 1.0, "step": 0.01}),
542
543
544
545
                             }}
    RETURN_TYPES = ("CONDITIONING",)
    FUNCTION = "apply_adm"

546
    CATEGORY = "conditioning"
547

548
    def apply_adm(self, conditioning, clip_vision_output, strength, noise_augmentation):
549
550
551
        c = []
        for t in conditioning:
            o = t[1].copy()
552
            x = (clip_vision_output, strength, noise_augmentation)
553
554
555
556
557
558
559
560
            if "adm" in o:
                o["adm"] = o["adm"][:] + [x]
            else:
                o["adm"] = [x]
            n = [t[0], o]
            c.append(n)
        return (c, )

561
562
563
564
565
566
567
568
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
569
    CATEGORY = "loaders"
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590

    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", ),
                              "text": ("STRING", {"multiline": True}),
                              "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
591
    CATEGORY = "conditioning/gligen"
592
593
594
595
596
597
598
599
600
601
602
603
604
605

    def append(self, conditioning_to, clip, gligen_textbox_model, text, width, height, x, y):
        c = []
        cond, cond_pooled = clip.encode_from_tokens(clip.tokenize(text), return_pooled=True)
        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, )
606

comfyanonymous's avatar
comfyanonymous committed
607
608
609
610
611
612
class EmptyLatentImage:
    def __init__(self, device="cpu"):
        self.device = device

    @classmethod
    def INPUT_TYPES(s):
613
614
        return {"required": { "width": ("INT", {"default": 512, "min": 64, "max": MAX_RESOLUTION, "step": 64}),
                              "height": ("INT", {"default": 512, "min": 64, "max": MAX_RESOLUTION, "step": 64}),
comfyanonymous's avatar
comfyanonymous committed
615
616
617
618
                              "batch_size": ("INT", {"default": 1, "min": 1, "max": 64})}}
    RETURN_TYPES = ("LATENT",)
    FUNCTION = "generate"

619
620
    CATEGORY = "latent"

comfyanonymous's avatar
comfyanonymous committed
621
622
    def generate(self, width, height, batch_size=1):
        latent = torch.zeros([batch_size, 4, height // 8, width // 8])
623
        return ({"samples":latent}, )
comfyanonymous's avatar
comfyanonymous committed
624

comfyanonymous's avatar
comfyanonymous committed
625

626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
class LatentFromBatch:
    @classmethod
    def INPUT_TYPES(s):
        return {"required": { "samples": ("LATENT",),
                              "batch_index": ("INT", {"default": 0, "min": 0, "max": 63}),
                              }}
    RETURN_TYPES = ("LATENT",)
    FUNCTION = "rotate"

    CATEGORY = "latent"

    def rotate(self, samples, batch_index):
        s = samples.copy()
        s_in = samples["samples"]
        batch_index = min(s_in.shape[0] - 1, batch_index)
        s["samples"] = s_in[batch_index:batch_index + 1].clone()
        s["batch_index"] = batch_index
        return (s,)
comfyanonymous's avatar
comfyanonymous committed
644

comfyanonymous's avatar
comfyanonymous committed
645
646
class LatentUpscale:
    upscale_methods = ["nearest-exact", "bilinear", "area"]
647
    crop_methods = ["disabled", "center"]
comfyanonymous's avatar
comfyanonymous committed
648
649
650
651

    @classmethod
    def INPUT_TYPES(s):
        return {"required": { "samples": ("LATENT",), "upscale_method": (s.upscale_methods,),
652
653
                              "width": ("INT", {"default": 512, "min": 64, "max": MAX_RESOLUTION, "step": 64}),
                              "height": ("INT", {"default": 512, "min": 64, "max": MAX_RESOLUTION, "step": 64}),
654
                              "crop": (s.crop_methods,)}}
comfyanonymous's avatar
comfyanonymous committed
655
656
657
    RETURN_TYPES = ("LATENT",)
    FUNCTION = "upscale"

658
659
    CATEGORY = "latent"

660
    def upscale(self, samples, upscale_method, width, height, crop):
661
        s = samples.copy()
comfyanonymous's avatar
comfyanonymous committed
662
        s["samples"] = comfy.utils.common_upscale(samples["samples"], width // 8, height // 8, upscale_method, crop)
comfyanonymous's avatar
comfyanonymous committed
663
664
        return (s,)

comfyanonymous's avatar
comfyanonymous committed
665
666
667
668
669
670
671
672
673
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
674
    CATEGORY = "latent/transform"
comfyanonymous's avatar
comfyanonymous committed
675
676

    def rotate(self, samples, rotation):
677
        s = samples.copy()
comfyanonymous's avatar
comfyanonymous committed
678
679
680
681
682
683
684
685
        rotate_by = 0
        if rotation.startswith("90"):
            rotate_by = 1
        elif rotation.startswith("180"):
            rotate_by = 2
        elif rotation.startswith("270"):
            rotate_by = 3

686
        s["samples"] = torch.rot90(samples["samples"], k=rotate_by, dims=[3, 2])
comfyanonymous's avatar
comfyanonymous committed
687
        return (s,)
comfyanonymous's avatar
comfyanonymous committed
688
689
690
691
692
693
694
695
696
697

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
698
    CATEGORY = "latent/transform"
comfyanonymous's avatar
comfyanonymous committed
699
700

    def flip(self, samples, flip_method):
701
        s = samples.copy()
comfyanonymous's avatar
comfyanonymous committed
702
        if flip_method.startswith("x"):
703
            s["samples"] = torch.flip(samples["samples"], dims=[2])
comfyanonymous's avatar
comfyanonymous committed
704
        elif flip_method.startswith("y"):
705
            s["samples"] = torch.flip(samples["samples"], dims=[3])
comfyanonymous's avatar
comfyanonymous committed
706
707

        return (s,)
comfyanonymous's avatar
comfyanonymous committed
708
709
710
711

class LatentComposite:
    @classmethod
    def INPUT_TYPES(s):
comfyanonymous's avatar
comfyanonymous committed
712
713
714
715
716
717
        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
718
719
720
721
722
    RETURN_TYPES = ("LATENT",)
    FUNCTION = "composite"

    CATEGORY = "latent"

comfyanonymous's avatar
comfyanonymous committed
723
724
725
    def composite(self, samples_to, samples_from, x, y, composite_method="normal", feather=0):
        x =  x // 8
        y = y // 8
726
        feather = feather // 8
comfyanonymous's avatar
comfyanonymous committed
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
        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
750

comfyanonymous's avatar
comfyanonymous committed
751
752
753
754
class LatentCrop:
    @classmethod
    def INPUT_TYPES(s):
        return {"required": { "samples": ("LATENT",),
755
756
757
758
                              "width": ("INT", {"default": 512, "min": 64, "max": MAX_RESOLUTION, "step": 64}),
                              "height": ("INT", {"default": 512, "min": 64, "max": MAX_RESOLUTION, "step": 64}),
                              "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
759
760
761
762
                              }}
    RETURN_TYPES = ("LATENT",)
    FUNCTION = "crop"

comfyanonymous's avatar
comfyanonymous committed
763
    CATEGORY = "latent/transform"
comfyanonymous's avatar
comfyanonymous committed
764
765

    def crop(self, samples, width, height, x, y):
766
767
        s = samples.copy()
        samples = samples['samples']
comfyanonymous's avatar
comfyanonymous committed
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
        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
        def enforce_image_dim(d, to_d, max_d):
            if to_d > max_d:
                leftover = (to_d - max_d) % 8
                to_d = max_d
                d -= leftover
            return (d, to_d)

        #make sure size is always multiple of 64
        x, to_x = enforce_image_dim(x, to_x, samples.shape[3])
        y, to_y = enforce_image_dim(y, to_y, samples.shape[2])
791
        s['samples'] = samples[:,:,y:to_y, x:to_x]
comfyanonymous's avatar
comfyanonymous committed
792
793
        return (s,)

794
795
796
797
798
799
800
801
802
class SetLatentNoiseMask:
    @classmethod
    def INPUT_TYPES(s):
        return {"required": { "samples": ("LATENT",),
                              "mask": ("MASK",),
                              }}
    RETURN_TYPES = ("LATENT",)
    FUNCTION = "set_mask"

803
    CATEGORY = "latent/inpaint"
804
805
806
807
808
809

    def set_mask(self, samples, mask):
        s = samples.copy()
        s["noise_mask"] = mask
        return (s,)

810
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):
811
    device = comfy.model_management.get_torch_device()
812
    latent_image = latent["samples"]
813

comfyanonymous's avatar
comfyanonymous committed
814
815
816
    if disable_noise:
        noise = torch.zeros(latent_image.size(), dtype=latent_image.dtype, layout=latent_image.layout, device="cpu")
    else:
817
818
        skip = latent["batch_index"] if "batch_index" in latent else 0
        noise = comfy.sample.prepare_noise(latent_image, seed, skip)
comfyanonymous's avatar
comfyanonymous committed
819

820
    noise_mask = None
821
    if "noise_mask" in latent:
822
        noise_mask = latent["noise_mask"]
comfyanonymous's avatar
comfyanonymous committed
823

824
825
826
    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,
                                  force_full_denoise=force_full_denoise, noise_mask=noise_mask)
827
828
829
    out = latent.copy()
    out["samples"] = samples
    return (out, )
comfyanonymous's avatar
comfyanonymous committed
830

comfyanonymous's avatar
comfyanonymous committed
831
832
833
class KSampler:
    @classmethod
    def INPUT_TYPES(s):
comfyanonymous's avatar
comfyanonymous committed
834
        return {"required":
comfyanonymous's avatar
comfyanonymous committed
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
                    {"model": ("MODEL",),
                    "seed": ("INT", {"default": 0, "min": 0, "max": 0xffffffffffffffff}),
                    "steps": ("INT", {"default": 20, "min": 1, "max": 10000}),
                    "cfg": ("FLOAT", {"default": 8.0, "min": 0.0, "max": 100.0}),
                    "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}),
                    }}

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

850
851
    CATEGORY = "sampling"

comfyanonymous's avatar
comfyanonymous committed
852
    def sample(self, model, seed, steps, cfg, sampler_name, scheduler, positive, negative, latent_image, denoise=1.0):
853
        return common_ksampler(model, seed, steps, cfg, sampler_name, scheduler, positive, negative, latent_image, denoise=denoise)
comfyanonymous's avatar
comfyanonymous committed
854

comfyanonymous's avatar
comfyanonymous committed
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
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}),
                    "cfg": ("FLOAT", {"default": 8.0, "min": 0.0, "max": 100.0}),
                    "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"], ),
                    }}

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

    CATEGORY = "sampling"
comfyanonymous's avatar
comfyanonymous committed
878

comfyanonymous's avatar
comfyanonymous committed
879
880
881
882
883
884
885
    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):
        force_full_denoise = True
        if return_with_leftover_noise == "enable":
            force_full_denoise = False
        disable_noise = False
        if add_noise == "disable":
            disable_noise = True
886
        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
887
888
889

class SaveImage:
    def __init__(self):
890
        self.output_dir = folder_paths.get_output_directory()
m957ymj75urz's avatar
m957ymj75urz committed
891
        self.type = "output"
comfyanonymous's avatar
comfyanonymous committed
892
893
894
895

    @classmethod
    def INPUT_TYPES(s):
        return {"required": 
896
                    {"images": ("IMAGE", ),
pythongosssss's avatar
tidy  
pythongosssss committed
897
                     "filename_prefix": ("STRING", {"default": "ComfyUI"})},
pythongosssss's avatar
pythongosssss committed
898
                "hidden": {"prompt": "PROMPT", "extra_pnginfo": "EXTRA_PNGINFO"},
comfyanonymous's avatar
comfyanonymous committed
899
900
901
902
903
904
905
                }

    RETURN_TYPES = ()
    FUNCTION = "save_images"

    OUTPUT_NODE = True

906
907
    CATEGORY = "image"

pythongosssss's avatar
tidy  
pythongosssss committed
908
    def save_images(self, images, filename_prefix="ComfyUI", prompt=None, extra_pnginfo=None):
909
        def map_filename(filename):
910
            prefix_len = len(os.path.basename(filename_prefix))
911
912
913
914
915
916
            prefix = filename[:prefix_len + 1]
            try:
                digits = int(filename[prefix_len + 1:].split('_')[0])
            except:
                digits = 0
            return (digits, prefix)
comfyanonymous's avatar
Style.  
comfyanonymous committed
917

918
919
920
921
        def compute_vars(input):
            input = input.replace("%width%", str(images[0].shape[1]))
            input = input.replace("%height%", str(images[0].shape[0]))
            return input
comfyanonymous's avatar
Style.  
comfyanonymous committed
922

923
        filename_prefix = compute_vars(filename_prefix)
comfyanonymous's avatar
comfyanonymous committed
924

m957ymj75urz's avatar
m957ymj75urz committed
925
926
927
        subfolder = os.path.dirname(os.path.normpath(filename_prefix))
        filename = os.path.basename(os.path.normpath(filename_prefix))

comfyanonymous's avatar
comfyanonymous committed
928
        full_output_folder = os.path.join(self.output_dir, subfolder)
929

930
        if os.path.commonpath((self.output_dir, os.path.abspath(full_output_folder))) != self.output_dir:
931
            print("Saving image outside the output folder is not allowed.")
comfyanonymous's avatar
comfyanonymous committed
932
933
            return {}

934
        try:
935
            counter = max(filter(lambda a: a[1][:-1] == filename and a[1][-1] == "_", map(map_filename, os.listdir(full_output_folder))))[0] + 1
936
937
        except ValueError:
            counter = 1
938
        except FileNotFoundError:
939
            os.makedirs(full_output_folder, exist_ok=True)
940
            counter = 1
pythongosssss's avatar
pythongosssss committed
941

m957ymj75urz's avatar
m957ymj75urz committed
942
        results = list()
comfyanonymous's avatar
comfyanonymous committed
943
944
        for image in images:
            i = 255. * image.cpu().numpy()
945
            img = Image.fromarray(np.clip(i, 0, 255).astype(np.uint8))
comfyanonymous's avatar
comfyanonymous committed
946
947
948
949
950
951
            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]))
952

953
            file = f"{filename}_{counter:05}_.png"
954
            img.save(os.path.join(full_output_folder, file), pnginfo=metadata, compress_level=4)
m957ymj75urz's avatar
m957ymj75urz committed
955
956
957
958
            results.append({
                "filename": file,
                "subfolder": subfolder,
                "type": self.type
Gavroche CryptoRUSH's avatar
Gavroche CryptoRUSH committed
959
            })
960
            counter += 1
comfyanonymous's avatar
comfyanonymous committed
961

m957ymj75urz's avatar
m957ymj75urz committed
962
        return { "ui": { "images": results } }
comfyanonymous's avatar
comfyanonymous committed
963

pythongosssss's avatar
pythongosssss committed
964
965
class PreviewImage(SaveImage):
    def __init__(self):
966
        self.output_dir = folder_paths.get_temp_directory()
m957ymj75urz's avatar
m957ymj75urz committed
967
        self.type = "temp"
pythongosssss's avatar
pythongosssss committed
968
969
970

    @classmethod
    def INPUT_TYPES(s):
971
        return {"required":
pythongosssss's avatar
pythongosssss committed
972
973
974
                    {"images": ("IMAGE", ), },
                "hidden": {"prompt": "PROMPT", "extra_pnginfo": "EXTRA_PNGINFO"},
                }
975

976
977
978
class LoadImage:
    @classmethod
    def INPUT_TYPES(s):
979
        input_dir = folder_paths.get_input_directory()
980
        return {"required":
981
                    {"image": (sorted(os.listdir(input_dir)), )},
982
                }
983
984

    CATEGORY = "image"
985

986
    RETURN_TYPES = ("IMAGE", "MASK")
987
988
    FUNCTION = "load_image"
    def load_image(self, image):
989
        image_path = folder_paths.get_annotated_filepath(image)
990
991
        i = Image.open(image_path)
        image = i.convert("RGB")
992
        image = np.array(image).astype(np.float32) / 255.0
993
        image = torch.from_numpy(image)[None,]
994
995
996
997
998
999
        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")
        return (image, mask)
1000

1001
1002
    @classmethod
    def IS_CHANGED(s, image):
1003
        image_path = folder_paths.get_annotated_filepath(image)
1004
1005
1006
1007
        m = hashlib.sha256()
        with open(image_path, 'rb') as f:
            m.update(f.read())
        return m.digest().hex()
pythongosssss's avatar
pythongosssss committed
1008

1009
1010
1011
1012
1013
1014
1015
    @classmethod
    def VALIDATE_INPUTS(s, image):
        if not folder_paths.exists_annotated_filepath(image):
            return "Invalid image file: {}".format(image)

        return True

1016
class LoadImageMask:
1017
    _color_channels = ["alpha", "red", "green", "blue"]
1018
1019
    @classmethod
    def INPUT_TYPES(s):
1020
        input_dir = folder_paths.get_input_directory()
1021
        return {"required":
1022
                    {"image": (sorted(os.listdir(input_dir)), ),
1023
                    "channel": (s._color_channels, ),}
1024
1025
                }

1026
    CATEGORY = "mask"
1027
1028
1029
1030

    RETURN_TYPES = ("MASK",)
    FUNCTION = "load_image"
    def load_image(self, image, channel):
1031
        image_path = folder_paths.get_annotated_filepath(image)
1032
        i = Image.open(image_path)
1033
1034
        if i.getbands() != ("R", "G", "B", "A"):
            i = i.convert("RGBA")
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
        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")
        return (mask,)

    @classmethod
    def IS_CHANGED(s, image, channel):
1048
        image_path = folder_paths.get_annotated_filepath(image)
1049
1050
1051
1052
        m = hashlib.sha256()
        with open(image_path, 'rb') as f:
            m.update(f.read())
        return m.digest().hex()
pythongosssss's avatar
pythongosssss committed
1053

1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
    @classmethod
    def VALIDATE_INPUTS(s, image, channel):
        if not folder_paths.exists_annotated_filepath(image):
            return "Invalid image file: {}".format(image)

        if channel not in s._color_channels:
            return "Invalid color channel: {}".format(channel)

        return True

comfyanonymous's avatar
comfyanonymous committed
1064
1065
1066
1067
1068
1069
1070
class ImageScale:
    upscale_methods = ["nearest-exact", "bilinear", "area"]
    crop_methods = ["disabled", "center"]

    @classmethod
    def INPUT_TYPES(s):
        return {"required": { "image": ("IMAGE",), "upscale_method": (s.upscale_methods,),
1071
1072
                              "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
1073
1074
1075
1076
                              "crop": (s.crop_methods,)}}
    RETURN_TYPES = ("IMAGE",)
    FUNCTION = "upscale"

1077
    CATEGORY = "image/upscaling"
1078

comfyanonymous's avatar
comfyanonymous committed
1079
1080
    def upscale(self, image, upscale_method, width, height, crop):
        samples = image.movedim(-1,1)
comfyanonymous's avatar
comfyanonymous committed
1081
        s = comfy.utils.common_upscale(samples, width, height, upscale_method, crop)
comfyanonymous's avatar
comfyanonymous committed
1082
1083
        s = s.movedim(1,-1)
        return (s,)
comfyanonymous's avatar
comfyanonymous committed
1084

1085
1086
1087
1088
1089
1090
1091
1092
1093
1094
1095
1096
1097
1098
1099
1100
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,)


Guo Y.K's avatar
Guo Y.K committed
1101
1102
1103
1104
1105
1106
1107
class ImagePadForOutpaint:

    @classmethod
    def INPUT_TYPES(s):
        return {
            "required": {
                "image": ("IMAGE",),
1108
1109
1110
1111
1112
                "left": ("INT", {"default": 0, "min": 0, "max": MAX_RESOLUTION, "step": 64}),
                "top": ("INT", {"default": 0, "min": 0, "max": MAX_RESOLUTION, "step": 64}),
                "right": ("INT", {"default": 0, "min": 0, "max": MAX_RESOLUTION, "step": 64}),
                "bottom": ("INT", {"default": 0, "min": 0, "max": MAX_RESOLUTION, "step": 64}),
                "feathering": ("INT", {"default": 40, "min": 0, "max": MAX_RESOLUTION, "step": 1}),
Guo Y.K's avatar
Guo Y.K committed
1113
1114
1115
1116
1117
1118
1119
1120
            }
        }

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

    CATEGORY = "image"

1121
    def expand_image(self, image, left, top, right, bottom, feathering):
Guo Y.K's avatar
Guo Y.K committed
1122
1123
1124
1125
1126
1127
1128
1129
1130
1131
1132
1133
        d1, d2, d3, d4 = image.size()

        new_image = torch.zeros(
            (d1, d2 + top + bottom, d3 + left + right, d4),
            dtype=torch.float32,
        )
        new_image[:, top:top + d2, left:left + d3, :] = image

        mask = torch.ones(
            (d2 + top + bottom, d3 + left + right),
            dtype=torch.float32,
        )
1134

1135
1136
1137
1138
1139
        t = torch.zeros(
            (d2, d3),
            dtype=torch.float32
        )

1140
        if feathering > 0 and feathering * 2 < d2 and feathering * 2 < d3:
1141
1142
1143
1144
1145
1146
1147
1148
1149
1150
1151
1152
1153
1154
1155
1156
1157
1158
1159

            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
1160

Guo Y.K's avatar
Guo Y.K committed
1161
1162
1163
        return (new_image, mask)


comfyanonymous's avatar
comfyanonymous committed
1164
1165
NODE_CLASS_MAPPINGS = {
    "KSampler": KSampler,
1166
    "CheckpointLoaderSimple": CheckpointLoaderSimple,
comfyanonymous's avatar
comfyanonymous committed
1167
    "CLIPTextEncode": CLIPTextEncode,
comfyanonymous's avatar
comfyanonymous committed
1168
    "CLIPSetLastLayer": CLIPSetLastLayer,
comfyanonymous's avatar
comfyanonymous committed
1169
1170
    "VAEDecode": VAEDecode,
    "VAEEncode": VAEEncode,
1171
    "VAEEncodeForInpaint": VAEEncodeForInpaint,
comfyanonymous's avatar
comfyanonymous committed
1172
1173
1174
    "VAELoader": VAELoader,
    "EmptyLatentImage": EmptyLatentImage,
    "LatentUpscale": LatentUpscale,
1175
    "LatentFromBatch": LatentFromBatch,
comfyanonymous's avatar
comfyanonymous committed
1176
    "SaveImage": SaveImage,
pythongosssss's avatar
pythongosssss committed
1177
    "PreviewImage": PreviewImage,
comfyanonymous's avatar
comfyanonymous committed
1178
    "LoadImage": LoadImage,
1179
    "LoadImageMask": LoadImageMask,
comfyanonymous's avatar
comfyanonymous committed
1180
    "ImageScale": ImageScale,
1181
    "ImageInvert": ImageInvert,
Guo Y.K's avatar
Guo Y.K committed
1182
    "ImagePadForOutpaint": ImagePadForOutpaint,
FizzleDorf's avatar
FizzleDorf committed
1183
    "ConditioningAverage ": ConditioningAverage ,
comfyanonymous's avatar
comfyanonymous committed
1184
1185
    "ConditioningCombine": ConditioningCombine,
    "ConditioningSetArea": ConditioningSetArea,
Jacob Segal's avatar
Jacob Segal committed
1186
    "ConditioningSetMask": ConditioningSetMask,
comfyanonymous's avatar
comfyanonymous committed
1187
    "KSamplerAdvanced": KSamplerAdvanced,
1188
    "SetLatentNoiseMask": SetLatentNoiseMask,
comfyanonymous's avatar
comfyanonymous committed
1189
    "LatentComposite": LatentComposite,
comfyanonymous's avatar
comfyanonymous committed
1190
    "LatentRotate": LatentRotate,
comfyanonymous's avatar
comfyanonymous committed
1191
    "LatentFlip": LatentFlip,
comfyanonymous's avatar
comfyanonymous committed
1192
    "LatentCrop": LatentCrop,
1193
    "LoraLoader": LoraLoader,
1194
    "CLIPLoader": CLIPLoader,
1195
    "CLIPVisionEncode": CLIPVisionEncode,
1196
    "StyleModelApply": StyleModelApply,
1197
    "unCLIPConditioning": unCLIPConditioning,
comfyanonymous's avatar
comfyanonymous committed
1198
1199
    "ControlNetApply": ControlNetApply,
    "ControlNetLoader": ControlNetLoader,
1200
    "DiffControlNetLoader": DiffControlNetLoader,
comfyanonymous's avatar
comfyanonymous committed
1201
1202
    "StyleModelLoader": StyleModelLoader,
    "CLIPVisionLoader": CLIPVisionLoader,
1203
    "VAEDecodeTiled": VAEDecodeTiled,
comfyanonymous's avatar
comfyanonymous committed
1204
    "VAEEncodeTiled": VAEEncodeTiled,
1205
    "TomePatchModel": TomePatchModel,
1206
    "unCLIPCheckpointLoader": unCLIPCheckpointLoader,
1207
1208
1209
    "GLIGENLoader": GLIGENLoader,
    "GLIGENTextBoxApply": GLIGENTextBoxApply,

1210
    "CheckpointLoader": CheckpointLoader,
sALTaccount's avatar
sALTaccount committed
1211
    "DiffusersLoader": DiffusersLoader,
comfyanonymous's avatar
comfyanonymous committed
1212
1213
}

City's avatar
City committed
1214
1215
1216
1217
1218
NODE_DISPLAY_NAME_MAPPINGS = {
    # Sampling
    "KSampler": "KSampler",
    "KSamplerAdvanced": "KSampler (Advanced)",
    # Loaders
comfyanonymous's avatar
Rename.  
comfyanonymous committed
1219
1220
    "CheckpointLoader": "Load Checkpoint (With Config)",
    "CheckpointLoaderSimple": "Load Checkpoint",
City's avatar
City committed
1221
1222
1223
1224
1225
1226
1227
1228
1229
1230
1231
1232
1233
1234
    "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
1235
    "ConditioningAverage ": "Conditioning (Average)",
City's avatar
City committed
1236
    "ConditioningSetArea": "Conditioning (Set Area)",
Jacob Segal's avatar
Jacob Segal committed
1237
    "ConditioningSetMask": "Conditioning (Set Mask)",
City's avatar
City committed
1238
1239
1240
1241
1242
1243
1244
1245
1246
1247
1248
1249
1250
1251
1252
1253
1254
1255
1256
1257
1258
1259
1260
1261
1262
1263
    "ControlNetApply": "Apply ControlNet",
    # 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",
    "LatentComposite": "Latent Composite",
    # Image
    "SaveImage": "Save Image",
    "PreviewImage": "Preview Image",
    "LoadImage": "Load Image",
    "LoadImageMask": "Load Image (as Mask)",
    "ImageScale": "Upscale Image",
    "ImageUpscaleWithModel": "Upscale Image (using Model)",
    "ImageInvert": "Invert Image",
    "ImagePadForOutpaint": "Pad Image for Outpainting",
    # _for_testing
    "VAEDecodeTiled": "VAE Decode (Tiled)",
    "VAEEncodeTiled": "VAE Encode (Tiled)",
}

1264
1265
1266
1267
1268
1269
1270
1271
1272
1273
1274
1275
1276
1277
1278
def load_custom_node(module_path):
    module_name = os.path.basename(module_path)
    if os.path.isfile(module_path):
        sp = os.path.splitext(module_path)
        module_name = sp[0]
    try:
        if os.path.isfile(module_path):
            module_spec = importlib.util.spec_from_file_location(module_name, module_path)
        else:
            module_spec = importlib.util.spec_from_file_location(module_name, os.path.join(module_path, "__init__.py"))
        module = importlib.util.module_from_spec(module_spec)
        sys.modules[module_name] = module
        module_spec.loader.exec_module(module)
        if hasattr(module, "NODE_CLASS_MAPPINGS") and getattr(module, "NODE_CLASS_MAPPINGS") is not None:
            NODE_CLASS_MAPPINGS.update(module.NODE_CLASS_MAPPINGS)
1279
1280
            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)
1281
1282
1283
1284
1285
1286
        else:
            print(f"Skip {module_path} module for custom nodes due to the lack of NODE_CLASS_MAPPINGS.")
    except Exception as e:
        print(traceback.format_exc())
        print(f"Cannot import {module_path} module for custom nodes:", e)

Hacker 17082006's avatar
Hacker 17082006 committed
1287
def load_custom_nodes():
1288
1289
1290
1291
1292
1293
1294
1295
1296
1297
    node_paths = folder_paths.get_folder_paths("custom_nodes")
    for custom_node_path in node_paths:
        possible_modules = os.listdir(custom_node_path)
        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
            load_custom_node(module_path)
1298

1299
1300
def init_custom_nodes():
    load_custom_nodes()
1301
    load_custom_node(os.path.join(os.path.join(os.path.dirname(os.path.realpath(__file__)), "comfy_extras"), "nodes_hypernetwork.py"))
1302
1303
    load_custom_node(os.path.join(os.path.join(os.path.dirname(os.path.realpath(__file__)), "comfy_extras"), "nodes_upscale_model.py"))
    load_custom_node(os.path.join(os.path.join(os.path.dirname(os.path.realpath(__file__)), "comfy_extras"), "nodes_post_processing.py"))
1304
    load_custom_node(os.path.join(os.path.join(os.path.dirname(os.path.realpath(__file__)), "comfy_extras"), "nodes_mask.py"))