Spaces:
Running
on
Zero
Running
on
Zero
File size: 68,204 Bytes
77f10a3 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 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 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229 1230 1231 1232 1233 1234 1235 1236 1237 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 1264 1265 1266 1267 1268 1269 1270 1271 1272 1273 1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292 1293 1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320 1321 1322 1323 1324 1325 1326 1327 1328 1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341 1342 1343 1344 1345 1346 1347 1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 1359 1360 1361 1362 1363 1364 1365 1366 1367 1368 1369 1370 1371 1372 1373 1374 1375 1376 1377 1378 1379 1380 1381 1382 1383 1384 1385 1386 1387 1388 1389 1390 1391 1392 1393 1394 1395 1396 1397 1398 1399 1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 1414 1415 1416 1417 1418 1419 1420 1421 1422 1423 1424 1425 1426 1427 1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 1439 1440 1441 1442 1443 1444 1445 1446 1447 1448 1449 1450 1451 1452 1453 1454 1455 1456 1457 1458 1459 1460 1461 1462 1463 1464 1465 1466 1467 1468 1469 1470 1471 1472 1473 1474 1475 1476 1477 1478 1479 1480 1481 1482 1483 1484 1485 1486 1487 1488 1489 1490 1491 1492 1493 1494 1495 1496 1497 1498 1499 1500 1501 1502 1503 1504 1505 1506 1507 1508 1509 1510 1511 1512 1513 1514 1515 1516 1517 1518 1519 1520 1521 1522 1523 1524 1525 1526 1527 1528 1529 1530 1531 1532 1533 1534 1535 1536 1537 1538 1539 1540 1541 1542 1543 1544 1545 1546 1547 1548 1549 1550 1551 1552 1553 1554 1555 1556 1557 1558 1559 1560 1561 1562 1563 1564 1565 1566 1567 1568 1569 1570 1571 1572 1573 1574 1575 1576 1577 1578 1579 1580 1581 1582 1583 1584 1585 1586 1587 1588 1589 1590 1591 1592 1593 1594 1595 1596 1597 1598 1599 1600 1601 1602 1603 1604 1605 1606 1607 1608 1609 1610 1611 1612 1613 1614 1615 1616 1617 1618 1619 1620 1621 1622 1623 1624 1625 1626 1627 1628 1629 1630 1631 1632 1633 1634 1635 1636 1637 1638 1639 1640 1641 1642 1643 1644 1645 1646 1647 1648 1649 1650 1651 1652 1653 1654 1655 1656 1657 1658 1659 1660 1661 1662 1663 1664 1665 1666 1667 1668 1669 1670 1671 1672 1673 1674 1675 1676 1677 1678 1679 1680 1681 1682 1683 1684 1685 1686 1687 1688 1689 1690 1691 1692 1693 1694 1695 1696 1697 1698 1699 1700 1701 1702 1703 1704 1705 1706 1707 1708 1709 1710 1711 1712 1713 1714 1715 1716 1717 1718 1719 1720 1721 1722 1723 1724 1725 1726 1727 1728 1729 1730 1731 1732 1733 1734 1735 1736 1737 1738 1739 1740 1741 1742 1743 1744 1745 1746 1747 1748 1749 1750 1751 1752 1753 1754 1755 1756 1757 1758 1759 |
"""Kling API Nodes
For source of truth on the allowed permutations of request fields, please reference:
- [Compatibility Table](https://app.klingai.com/global/dev/document-api/apiReference/model/skillsMap)
"""
from __future__ import annotations
from typing import Optional, TypeVar, Any
from collections.abc import Callable
import math
import logging
import torch
from comfy_api_nodes.apis import (
KlingTaskStatus,
KlingCameraControl,
KlingCameraConfig,
KlingCameraControlType,
KlingVideoGenDuration,
KlingVideoGenMode,
KlingVideoGenAspectRatio,
KlingVideoGenModelName,
KlingText2VideoRequest,
KlingText2VideoResponse,
KlingImage2VideoRequest,
KlingImage2VideoResponse,
KlingVideoExtendRequest,
KlingVideoExtendResponse,
KlingLipSyncVoiceLanguage,
KlingLipSyncInputObject,
KlingLipSyncRequest,
KlingLipSyncResponse,
KlingVirtualTryOnModelName,
KlingVirtualTryOnRequest,
KlingVirtualTryOnResponse,
KlingVideoResult,
KlingImageResult,
KlingImageGenerationsRequest,
KlingImageGenerationsResponse,
KlingImageGenImageReferenceType,
KlingImageGenModelName,
KlingImageGenAspectRatio,
KlingVideoEffectsRequest,
KlingVideoEffectsResponse,
KlingDualCharacterEffectsScene,
KlingSingleImageEffectsScene,
KlingDualCharacterEffectInput,
KlingSingleImageEffectInput,
KlingCharacterEffectModelName,
KlingSingleImageEffectModelName,
)
from comfy_api_nodes.apis.client import (
ApiEndpoint,
HttpMethod,
SynchronousOperation,
PollingOperation,
EmptyRequest,
)
from comfy_api_nodes.apinode_utils import (
tensor_to_base64_string,
download_url_to_video_output,
upload_video_to_comfyapi,
upload_audio_to_comfyapi,
download_url_to_image_tensor,
)
from comfy_api_nodes.mapper_utils import model_field_to_node_input
from comfy_api_nodes.util.validation_utils import (
validate_image_dimensions,
validate_image_aspect_ratio,
validate_video_dimensions,
validate_video_duration,
)
from comfy_api.input.basic_types import AudioInput
from comfy_api.input.video_types import VideoInput
from comfy_api.input_impl import VideoFromFile
from comfy.comfy_types.node_typing import IO, InputTypeOptions, ComfyNodeABC
KLING_API_VERSION = "v1"
PATH_TEXT_TO_VIDEO = f"/proxy/kling/{KLING_API_VERSION}/videos/text2video"
PATH_IMAGE_TO_VIDEO = f"/proxy/kling/{KLING_API_VERSION}/videos/image2video"
PATH_VIDEO_EXTEND = f"/proxy/kling/{KLING_API_VERSION}/videos/video-extend"
PATH_LIP_SYNC = f"/proxy/kling/{KLING_API_VERSION}/videos/lip-sync"
PATH_VIDEO_EFFECTS = f"/proxy/kling/{KLING_API_VERSION}/videos/effects"
PATH_CHARACTER_IMAGE = f"/proxy/kling/{KLING_API_VERSION}/images/generations"
PATH_VIRTUAL_TRY_ON = f"/proxy/kling/{KLING_API_VERSION}/images/kolors-virtual-try-on"
PATH_IMAGE_GENERATIONS = f"/proxy/kling/{KLING_API_VERSION}/images/generations"
MAX_PROMPT_LENGTH_T2V = 2500
MAX_PROMPT_LENGTH_I2V = 500
MAX_PROMPT_LENGTH_IMAGE_GEN = 500
MAX_NEGATIVE_PROMPT_LENGTH_IMAGE_GEN = 200
MAX_PROMPT_LENGTH_LIP_SYNC = 120
AVERAGE_DURATION_T2V = 319
AVERAGE_DURATION_I2V = 164
AVERAGE_DURATION_LIP_SYNC = 455
AVERAGE_DURATION_VIRTUAL_TRY_ON = 19
AVERAGE_DURATION_IMAGE_GEN = 32
AVERAGE_DURATION_VIDEO_EFFECTS = 320
AVERAGE_DURATION_VIDEO_EXTEND = 320
R = TypeVar("R")
class KlingApiError(Exception):
"""Base exception for Kling API errors."""
pass
def poll_until_finished(
auth_kwargs: dict[str, str],
api_endpoint: ApiEndpoint[Any, R],
result_url_extractor: Optional[Callable[[R], str]] = None,
estimated_duration: Optional[int] = None,
node_id: Optional[str] = None,
) -> R:
"""Polls the Kling API endpoint until the task reaches a terminal state, then returns the response."""
return PollingOperation(
poll_endpoint=api_endpoint,
completed_statuses=[
KlingTaskStatus.succeed.value,
],
failed_statuses=[KlingTaskStatus.failed.value],
status_extractor=lambda response: (
response.data.task_status.value
if response.data and response.data.task_status
else None
),
auth_kwargs=auth_kwargs,
result_url_extractor=result_url_extractor,
estimated_duration=estimated_duration,
node_id=node_id,
).execute()
def is_valid_camera_control_configs(configs: list[float]) -> bool:
"""Verifies that at least one camera control configuration is non-zero."""
return any(not math.isclose(value, 0.0) for value in configs)
def is_valid_prompt(prompt: str) -> bool:
"""Verifies that the prompt is not empty."""
return bool(prompt)
def is_valid_task_creation_response(response: KlingText2VideoResponse) -> bool:
"""Verifies that the initial response contains a task ID."""
return bool(response.data.task_id)
def is_valid_video_response(response: KlingText2VideoResponse) -> bool:
"""Verifies that the response contains a task result with at least one video."""
return (
response.data is not None
and response.data.task_result is not None
and response.data.task_result.videos is not None
and len(response.data.task_result.videos) > 0
)
def is_valid_image_response(response: KlingVirtualTryOnResponse) -> bool:
"""Verifies that the response contains a task result with at least one image."""
return (
response.data is not None
and response.data.task_result is not None
and response.data.task_result.images is not None
and len(response.data.task_result.images) > 0
)
def validate_prompts(prompt: str, negative_prompt: str, max_length: int) -> bool:
"""Verifies that the positive prompt is not empty and that neither promt is too long."""
if not prompt:
raise ValueError("Positive prompt is empty")
if len(prompt) > max_length:
raise ValueError(f"Positive prompt is too long: {len(prompt)} characters")
if negative_prompt and len(negative_prompt) > max_length:
raise ValueError(
f"Negative prompt is too long: {len(negative_prompt)} characters"
)
return True
def validate_task_creation_response(response) -> None:
"""Validates that the Kling task creation request was successful."""
if not is_valid_task_creation_response(response):
error_msg = f"Kling initial request failed. Code: {response.code}, Message: {response.message}, Data: {response.data}"
logging.error(error_msg)
raise KlingApiError(error_msg)
def validate_video_result_response(response) -> None:
"""Validates that the Kling task result contains a video."""
if not is_valid_video_response(response):
error_msg = f"Kling task {response.data.task_id} succeeded but no video data found in response."
logging.error(f"Error: {error_msg}.\nResponse: {response}")
raise KlingApiError(error_msg)
def validate_image_result_response(response) -> None:
"""Validates that the Kling task result contains an image."""
if not is_valid_image_response(response):
error_msg = f"Kling task {response.data.task_id} succeeded but no image data found in response."
logging.error(f"Error: {error_msg}.\nResponse: {response}")
raise KlingApiError(error_msg)
def validate_input_image(image: torch.Tensor) -> None:
"""
Validates the input image adheres to the expectations of the Kling API:
- The image resolution should not be less than 300*300px
- The aspect ratio of the image should be between 1:2.5 ~ 2.5:1
See: https://app.klingai.com/global/dev/document-api/apiReference/model/imageToVideo
"""
validate_image_dimensions(image, min_width=300, min_height=300)
validate_image_aspect_ratio(image, min_aspect_ratio=1 / 2.5, max_aspect_ratio=2.5)
def get_camera_control_input_config(
tooltip: str, default: float = 0.0
) -> tuple[IO, InputTypeOptions]:
"""Returns common InputTypeOptions for Kling camera control configurations."""
input_config = {
"default": default,
"min": -10.0,
"max": 10.0,
"step": 0.25,
"display": "slider",
"tooltip": tooltip,
}
return IO.FLOAT, input_config
def get_video_from_response(response) -> KlingVideoResult:
"""Returns the first video object from the Kling video generation task result.
Will raise an error if the response is not valid.
"""
video = response.data.task_result.videos[0]
logging.info(
"Kling task %s succeeded. Video URL: %s", response.data.task_id, video.url
)
return video
def get_video_url_from_response(response) -> Optional[str]:
"""Returns the first video url from the Kling video generation task result.
Will not raise an error if the response is not valid.
"""
if response and is_valid_video_response(response):
return str(get_video_from_response(response).url)
else:
return None
def get_images_from_response(response) -> list[KlingImageResult]:
"""Returns the list of image objects from the Kling image generation task result.
Will raise an error if the response is not valid.
"""
images = response.data.task_result.images
logging.info("Kling task %s succeeded. Images: %s", response.data.task_id, images)
return images
def get_images_urls_from_response(response) -> Optional[str]:
"""Returns the list of image urls from the Kling image generation task result.
Will not raise an error if the response is not valid. If there is only one image, returns the url as a string. If there are multiple images, returns a list of urls.
"""
if response and is_valid_image_response(response):
images = get_images_from_response(response)
image_urls = [str(image.url) for image in images]
return "\n".join(image_urls)
else:
return None
def video_result_to_node_output(
video: KlingVideoResult,
) -> tuple[VideoFromFile, str, str]:
"""Converts a KlingVideoResult to a tuple of (VideoFromFile, str, str) to be used as a ComfyUI node output."""
return (
download_url_to_video_output(video.url),
str(video.id),
str(video.duration),
)
def image_result_to_node_output(
images: list[KlingImageResult],
) -> torch.Tensor:
"""
Converts a KlingImageResult to a tuple containing a [B, H, W, C] tensor.
If multiple images are returned, they will be stacked along the batch dimension.
"""
if len(images) == 1:
return download_url_to_image_tensor(images[0].url)
else:
return torch.cat([download_url_to_image_tensor(image.url) for image in images])
class KlingNodeBase(ComfyNodeABC):
"""Base class for Kling nodes."""
FUNCTION = "api_call"
CATEGORY = "api node/video/Kling"
API_NODE = True
class KlingCameraControls(KlingNodeBase):
"""Kling Camera Controls Node"""
@classmethod
def INPUT_TYPES(cls):
return {
"required": {
"camera_control_type": model_field_to_node_input(
IO.COMBO,
KlingCameraControl,
"type",
enum_type=KlingCameraControlType,
),
"horizontal_movement": get_camera_control_input_config(
"Controls camera's movement along horizontal axis (x-axis). Negative indicates left, positive indicates right"
),
"vertical_movement": get_camera_control_input_config(
"Controls camera's movement along vertical axis (y-axis). Negative indicates downward, positive indicates upward."
),
"pan": get_camera_control_input_config(
"Controls camera's rotation in vertical plane (x-axis). Negative indicates downward rotation, positive indicates upward rotation.",
default=0.5,
),
"tilt": get_camera_control_input_config(
"Controls camera's rotation in horizontal plane (y-axis). Negative indicates left rotation, positive indicates right rotation.",
),
"roll": get_camera_control_input_config(
"Controls camera's rolling amount (z-axis). Negative indicates counterclockwise, positive indicates clockwise.",
),
"zoom": get_camera_control_input_config(
"Controls change in camera's focal length. Negative indicates narrower field of view, positive indicates wider field of view.",
),
}
}
DESCRIPTION = "Allows specifying configuration options for Kling Camera Controls and motion control effects."
RETURN_TYPES = ("CAMERA_CONTROL",)
RETURN_NAMES = ("camera_control",)
FUNCTION = "main"
API_NODE = False # This is just a helper node, it doesn't make an API call
@classmethod
def VALIDATE_INPUTS(
cls,
horizontal_movement: float,
vertical_movement: float,
pan: float,
tilt: float,
roll: float,
zoom: float,
) -> bool | str:
if not is_valid_camera_control_configs(
[
horizontal_movement,
vertical_movement,
pan,
tilt,
roll,
zoom,
]
):
return "Invalid camera control configs: at least one of the values must be non-zero"
return True
def main(
self,
camera_control_type: str,
horizontal_movement: float,
vertical_movement: float,
pan: float,
tilt: float,
roll: float,
zoom: float,
) -> tuple[KlingCameraControl]:
return (
KlingCameraControl(
type=KlingCameraControlType(camera_control_type),
config=KlingCameraConfig(
horizontal=horizontal_movement,
vertical=vertical_movement,
pan=pan,
roll=roll,
tilt=tilt,
zoom=zoom,
),
),
)
class KlingTextToVideoNode(KlingNodeBase):
"""Kling Text to Video Node"""
@staticmethod
def get_mode_string_mapping() -> dict[str, tuple[str, str, str]]:
"""
Returns a mapping of mode strings to their corresponding (mode, duration, model_name) tuples.
Only includes config combos that support the `image_tail` request field.
See: [Kling API Docs Capability Map](https://app.klingai.com/global/dev/document-api/apiReference/model/skillsMap)
"""
return {
"standard mode / 5s duration / kling-v1": ("std", "5", "kling-v1"),
"standard mode / 10s duration / kling-v1": ("std", "10", "kling-v1"),
"pro mode / 5s duration / kling-v1": ("pro", "5", "kling-v1"),
"pro mode / 10s duration / kling-v1": ("pro", "10", "kling-v1"),
"standard mode / 5s duration / kling-v1-6": ("std", "5", "kling-v1-6"),
"standard mode / 10s duration / kling-v1-6": ("std", "10", "kling-v1-6"),
"pro mode / 5s duration / kling-v2-master": ("pro", "5", "kling-v2-master"),
"pro mode / 10s duration / kling-v2-master": ("pro", "10", "kling-v2-master"),
"standard mode / 5s duration / kling-v2-master": ("std", "5", "kling-v2-master"),
"standard mode / 10s duration / kling-v2-master": ("std", "10", "kling-v2-master"),
}
@classmethod
def INPUT_TYPES(s):
modes = list(KlingTextToVideoNode.get_mode_string_mapping().keys())
return {
"required": {
"prompt": model_field_to_node_input(
IO.STRING, KlingText2VideoRequest, "prompt", multiline=True
),
"negative_prompt": model_field_to_node_input(
IO.STRING, KlingText2VideoRequest, "negative_prompt", multiline=True
),
"cfg_scale": model_field_to_node_input(
IO.FLOAT,
KlingText2VideoRequest,
"cfg_scale",
default=1.0,
min=0.0,
max=1.0,
),
"aspect_ratio": model_field_to_node_input(
IO.COMBO,
KlingText2VideoRequest,
"aspect_ratio",
enum_type=KlingVideoGenAspectRatio,
),
"mode": (
modes,
{
"default": modes[4],
"tooltip": "The configuration to use for the video generation following the format: mode / duration / model_name.",
},
),
},
"hidden": {
"auth_token": "AUTH_TOKEN_COMFY_ORG",
"comfy_api_key": "API_KEY_COMFY_ORG",
"unique_id": "UNIQUE_ID",
},
}
RETURN_TYPES = ("VIDEO", "STRING", "STRING")
RETURN_NAMES = ("VIDEO", "video_id", "duration")
DESCRIPTION = "Kling Text to Video Node"
def get_response(
self, task_id: str, auth_kwargs: dict[str, str], node_id: Optional[str] = None
) -> KlingText2VideoResponse:
return poll_until_finished(
auth_kwargs,
ApiEndpoint(
path=f"{PATH_TEXT_TO_VIDEO}/{task_id}",
method=HttpMethod.GET,
request_model=EmptyRequest,
response_model=KlingText2VideoResponse,
),
result_url_extractor=get_video_url_from_response,
estimated_duration=AVERAGE_DURATION_T2V,
node_id=node_id,
)
def api_call(
self,
prompt: str,
negative_prompt: str,
cfg_scale: float,
mode: str,
aspect_ratio: str,
camera_control: Optional[KlingCameraControl] = None,
model_name: Optional[str] = None,
duration: Optional[str] = None,
unique_id: Optional[str] = None,
**kwargs,
) -> tuple[VideoFromFile, str, str]:
validate_prompts(prompt, negative_prompt, MAX_PROMPT_LENGTH_T2V)
if model_name is None:
mode, duration, model_name = self.get_mode_string_mapping()[mode]
initial_operation = SynchronousOperation(
endpoint=ApiEndpoint(
path=PATH_TEXT_TO_VIDEO,
method=HttpMethod.POST,
request_model=KlingText2VideoRequest,
response_model=KlingText2VideoResponse,
),
request=KlingText2VideoRequest(
prompt=prompt if prompt else None,
negative_prompt=negative_prompt if negative_prompt else None,
duration=KlingVideoGenDuration(duration),
mode=KlingVideoGenMode(mode),
model_name=KlingVideoGenModelName(model_name),
cfg_scale=cfg_scale,
aspect_ratio=KlingVideoGenAspectRatio(aspect_ratio),
camera_control=camera_control,
),
auth_kwargs=kwargs,
)
task_creation_response = initial_operation.execute()
validate_task_creation_response(task_creation_response)
task_id = task_creation_response.data.task_id
final_response = self.get_response(
task_id, auth_kwargs=kwargs, node_id=unique_id
)
validate_video_result_response(final_response)
video = get_video_from_response(final_response)
return video_result_to_node_output(video)
class KlingCameraControlT2VNode(KlingTextToVideoNode):
"""
Kling Text to Video Camera Control Node. This node is a text to video node, but it supports controlling the camera.
Duration, mode, and model_name request fields are hard-coded because camera control is only supported in pro mode with the kling-v1-5 model at 5s duration as of 2025-05-02.
"""
@classmethod
def INPUT_TYPES(s):
return {
"required": {
"prompt": model_field_to_node_input(
IO.STRING, KlingText2VideoRequest, "prompt", multiline=True
),
"negative_prompt": model_field_to_node_input(
IO.STRING,
KlingText2VideoRequest,
"negative_prompt",
multiline=True,
),
"cfg_scale": model_field_to_node_input(
IO.FLOAT,
KlingText2VideoRequest,
"cfg_scale",
default=0.75,
min=0.0,
max=1.0,
),
"aspect_ratio": model_field_to_node_input(
IO.COMBO,
KlingText2VideoRequest,
"aspect_ratio",
enum_type=KlingVideoGenAspectRatio,
),
"camera_control": (
"CAMERA_CONTROL",
{
"tooltip": "Can be created using the Kling Camera Controls node. Controls the camera movement and motion during the video generation.",
},
),
},
"hidden": {
"auth_token": "AUTH_TOKEN_COMFY_ORG",
"comfy_api_key": "API_KEY_COMFY_ORG",
"unique_id": "UNIQUE_ID",
},
}
DESCRIPTION = "Transform text into cinematic videos with professional camera movements that simulate real-world cinematography. Control virtual camera actions including zoom, rotation, pan, tilt, and first-person view, while maintaining focus on your original text."
def api_call(
self,
prompt: str,
negative_prompt: str,
cfg_scale: float,
aspect_ratio: str,
camera_control: Optional[KlingCameraControl] = None,
unique_id: Optional[str] = None,
**kwargs,
):
return super().api_call(
model_name=KlingVideoGenModelName.kling_v1,
cfg_scale=cfg_scale,
mode=KlingVideoGenMode.std,
aspect_ratio=KlingVideoGenAspectRatio(aspect_ratio),
duration=KlingVideoGenDuration.field_5,
prompt=prompt,
negative_prompt=negative_prompt,
camera_control=camera_control,
**kwargs,
)
class KlingImage2VideoNode(KlingNodeBase):
"""Kling Image to Video Node"""
@classmethod
def INPUT_TYPES(s):
return {
"required": {
"start_frame": model_field_to_node_input(
IO.IMAGE,
KlingImage2VideoRequest,
"image",
tooltip="The reference image used to generate the video.",
),
"prompt": model_field_to_node_input(
IO.STRING, KlingImage2VideoRequest, "prompt", multiline=True
),
"negative_prompt": model_field_to_node_input(
IO.STRING,
KlingImage2VideoRequest,
"negative_prompt",
multiline=True,
),
"model_name": model_field_to_node_input(
IO.COMBO,
KlingImage2VideoRequest,
"model_name",
enum_type=KlingVideoGenModelName,
),
"cfg_scale": model_field_to_node_input(
IO.FLOAT,
KlingImage2VideoRequest,
"cfg_scale",
default=0.8,
min=0.0,
max=1.0,
),
"mode": model_field_to_node_input(
IO.COMBO,
KlingImage2VideoRequest,
"mode",
enum_type=KlingVideoGenMode,
),
"aspect_ratio": model_field_to_node_input(
IO.COMBO,
KlingImage2VideoRequest,
"aspect_ratio",
enum_type=KlingVideoGenAspectRatio,
),
"duration": model_field_to_node_input(
IO.COMBO,
KlingImage2VideoRequest,
"duration",
enum_type=KlingVideoGenDuration,
),
},
"hidden": {
"auth_token": "AUTH_TOKEN_COMFY_ORG",
"comfy_api_key": "API_KEY_COMFY_ORG",
"unique_id": "UNIQUE_ID",
},
}
RETURN_TYPES = ("VIDEO", "STRING", "STRING")
RETURN_NAMES = ("VIDEO", "video_id", "duration")
DESCRIPTION = "Kling Image to Video Node"
def get_response(
self, task_id: str, auth_kwargs: dict[str, str], node_id: Optional[str] = None
) -> KlingImage2VideoResponse:
return poll_until_finished(
auth_kwargs,
ApiEndpoint(
path=f"{PATH_IMAGE_TO_VIDEO}/{task_id}",
method=HttpMethod.GET,
request_model=KlingImage2VideoRequest,
response_model=KlingImage2VideoResponse,
),
result_url_extractor=get_video_url_from_response,
estimated_duration=AVERAGE_DURATION_I2V,
node_id=node_id,
)
def api_call(
self,
start_frame: torch.Tensor,
prompt: str,
negative_prompt: str,
model_name: str,
cfg_scale: float,
mode: str,
aspect_ratio: str,
duration: str,
camera_control: Optional[KlingCameraControl] = None,
end_frame: Optional[torch.Tensor] = None,
unique_id: Optional[str] = None,
**kwargs,
) -> tuple[VideoFromFile]:
validate_prompts(prompt, negative_prompt, MAX_PROMPT_LENGTH_I2V)
validate_input_image(start_frame)
if camera_control is not None:
# Camera control type for image 2 video is always `simple`
camera_control.type = KlingCameraControlType.simple
initial_operation = SynchronousOperation(
endpoint=ApiEndpoint(
path=PATH_IMAGE_TO_VIDEO,
method=HttpMethod.POST,
request_model=KlingImage2VideoRequest,
response_model=KlingImage2VideoResponse,
),
request=KlingImage2VideoRequest(
model_name=KlingVideoGenModelName(model_name),
image=tensor_to_base64_string(start_frame),
image_tail=(
tensor_to_base64_string(end_frame)
if end_frame is not None
else None
),
prompt=prompt,
negative_prompt=negative_prompt if negative_prompt else None,
cfg_scale=cfg_scale,
mode=KlingVideoGenMode(mode),
duration=KlingVideoGenDuration(duration),
camera_control=camera_control,
),
auth_kwargs=kwargs,
)
task_creation_response = initial_operation.execute()
validate_task_creation_response(task_creation_response)
task_id = task_creation_response.data.task_id
final_response = self.get_response(
task_id, auth_kwargs=kwargs, node_id=unique_id
)
validate_video_result_response(final_response)
video = get_video_from_response(final_response)
return video_result_to_node_output(video)
class KlingCameraControlI2VNode(KlingImage2VideoNode):
"""
Kling Image to Video Camera Control Node. This node is a image to video node, but it supports controlling the camera.
Duration, mode, and model_name request fields are hard-coded because camera control is only supported in pro mode with the kling-v1-5 model at 5s duration as of 2025-05-02.
"""
@classmethod
def INPUT_TYPES(s):
return {
"required": {
"start_frame": model_field_to_node_input(
IO.IMAGE, KlingImage2VideoRequest, "image"
),
"prompt": model_field_to_node_input(
IO.STRING, KlingImage2VideoRequest, "prompt", multiline=True
),
"negative_prompt": model_field_to_node_input(
IO.STRING,
KlingImage2VideoRequest,
"negative_prompt",
multiline=True,
),
"cfg_scale": model_field_to_node_input(
IO.FLOAT,
KlingImage2VideoRequest,
"cfg_scale",
default=0.75,
min=0.0,
max=1.0,
),
"aspect_ratio": model_field_to_node_input(
IO.COMBO,
KlingImage2VideoRequest,
"aspect_ratio",
enum_type=KlingVideoGenAspectRatio,
),
"camera_control": (
"CAMERA_CONTROL",
{
"tooltip": "Can be created using the Kling Camera Controls node. Controls the camera movement and motion during the video generation.",
},
),
},
"hidden": {
"auth_token": "AUTH_TOKEN_COMFY_ORG",
"comfy_api_key": "API_KEY_COMFY_ORG",
"unique_id": "UNIQUE_ID",
},
}
DESCRIPTION = "Transform still images into cinematic videos with professional camera movements that simulate real-world cinematography. Control virtual camera actions including zoom, rotation, pan, tilt, and first-person view, while maintaining focus on your original image."
def api_call(
self,
start_frame: torch.Tensor,
prompt: str,
negative_prompt: str,
cfg_scale: float,
aspect_ratio: str,
camera_control: KlingCameraControl,
unique_id: Optional[str] = None,
**kwargs,
):
return super().api_call(
model_name=KlingVideoGenModelName.kling_v1_5,
start_frame=start_frame,
cfg_scale=cfg_scale,
mode=KlingVideoGenMode.pro,
aspect_ratio=KlingVideoGenAspectRatio(aspect_ratio),
duration=KlingVideoGenDuration.field_5,
prompt=prompt,
negative_prompt=negative_prompt,
camera_control=camera_control,
unique_id=unique_id,
**kwargs,
)
class KlingStartEndFrameNode(KlingImage2VideoNode):
"""
Kling First Last Frame Node. This node allows creation of a video from a first and last frame. It calls the normal image to video endpoint, but only allows the subset of input options that support the `image_tail` request field.
"""
@staticmethod
def get_mode_string_mapping() -> dict[str, tuple[str, str, str]]:
"""
Returns a mapping of mode strings to their corresponding (mode, duration, model_name) tuples.
Only includes config combos that support the `image_tail` request field.
See: [Kling API Docs Capability Map](https://app.klingai.com/global/dev/document-api/apiReference/model/skillsMap)
"""
return {
"standard mode / 5s duration / kling-v1": ("std", "5", "kling-v1"),
"pro mode / 5s duration / kling-v1": ("pro", "5", "kling-v1"),
"pro mode / 5s duration / kling-v1-5": ("pro", "5", "kling-v1-5"),
"pro mode / 10s duration / kling-v1-5": ("pro", "10", "kling-v1-5"),
"pro mode / 5s duration / kling-v1-6": ("pro", "5", "kling-v1-6"),
"pro mode / 10s duration / kling-v1-6": ("pro", "10", "kling-v1-6"),
}
@classmethod
def INPUT_TYPES(s):
modes = list(KlingStartEndFrameNode.get_mode_string_mapping().keys())
return {
"required": {
"start_frame": model_field_to_node_input(
IO.IMAGE, KlingImage2VideoRequest, "image"
),
"end_frame": model_field_to_node_input(
IO.IMAGE, KlingImage2VideoRequest, "image_tail"
),
"prompt": model_field_to_node_input(
IO.STRING, KlingImage2VideoRequest, "prompt", multiline=True
),
"negative_prompt": model_field_to_node_input(
IO.STRING,
KlingImage2VideoRequest,
"negative_prompt",
multiline=True,
),
"cfg_scale": model_field_to_node_input(
IO.FLOAT,
KlingImage2VideoRequest,
"cfg_scale",
default=0.5,
min=0.0,
max=1.0,
),
"aspect_ratio": model_field_to_node_input(
IO.COMBO,
KlingImage2VideoRequest,
"aspect_ratio",
enum_type=KlingVideoGenAspectRatio,
),
"mode": (
modes,
{
"default": modes[2],
"tooltip": "The configuration to use for the video generation following the format: mode / duration / model_name.",
},
),
},
"hidden": {
"auth_token": "AUTH_TOKEN_COMFY_ORG",
"comfy_api_key": "API_KEY_COMFY_ORG",
"unique_id": "UNIQUE_ID",
},
}
DESCRIPTION = "Generate a video sequence that transitions between your provided start and end images. The node creates all frames in between, producing a smooth transformation from the first frame to the last."
def api_call(
self,
start_frame: torch.Tensor,
end_frame: torch.Tensor,
prompt: str,
negative_prompt: str,
cfg_scale: float,
aspect_ratio: str,
mode: str,
unique_id: Optional[str] = None,
**kwargs,
):
mode, duration, model_name = KlingStartEndFrameNode.get_mode_string_mapping()[
mode
]
return super().api_call(
prompt=prompt,
negative_prompt=negative_prompt,
model_name=model_name,
start_frame=start_frame,
cfg_scale=cfg_scale,
mode=mode,
aspect_ratio=aspect_ratio,
duration=duration,
end_frame=end_frame,
unique_id=unique_id,
**kwargs,
)
class KlingVideoExtendNode(KlingNodeBase):
@classmethod
def INPUT_TYPES(s):
return {
"required": {
"prompt": model_field_to_node_input(
IO.STRING, KlingVideoExtendRequest, "prompt", multiline=True
),
"negative_prompt": model_field_to_node_input(
IO.STRING,
KlingVideoExtendRequest,
"negative_prompt",
multiline=True,
),
"cfg_scale": model_field_to_node_input(
IO.FLOAT,
KlingVideoExtendRequest,
"cfg_scale",
default=0.5,
min=0.0,
max=1.0,
),
"video_id": model_field_to_node_input(
IO.STRING, KlingVideoExtendRequest, "video_id", forceInput=True
),
},
"hidden": {
"auth_token": "AUTH_TOKEN_COMFY_ORG",
"comfy_api_key": "API_KEY_COMFY_ORG",
"unique_id": "UNIQUE_ID",
},
}
RETURN_TYPES = ("VIDEO", "STRING", "STRING")
RETURN_NAMES = ("VIDEO", "video_id", "duration")
DESCRIPTION = "Kling Video Extend Node. Extend videos made by other Kling nodes. The video_id is created by using other Kling Nodes."
def get_response(
self, task_id: str, auth_kwargs: dict[str, str], node_id: Optional[str] = None
) -> KlingVideoExtendResponse:
return poll_until_finished(
auth_kwargs,
ApiEndpoint(
path=f"{PATH_VIDEO_EXTEND}/{task_id}",
method=HttpMethod.GET,
request_model=EmptyRequest,
response_model=KlingVideoExtendResponse,
),
result_url_extractor=get_video_url_from_response,
estimated_duration=AVERAGE_DURATION_VIDEO_EXTEND,
node_id=node_id,
)
def api_call(
self,
prompt: str,
negative_prompt: str,
cfg_scale: float,
video_id: str,
unique_id: Optional[str] = None,
**kwargs,
) -> tuple[VideoFromFile, str, str]:
validate_prompts(prompt, negative_prompt, MAX_PROMPT_LENGTH_T2V)
initial_operation = SynchronousOperation(
endpoint=ApiEndpoint(
path=PATH_VIDEO_EXTEND,
method=HttpMethod.POST,
request_model=KlingVideoExtendRequest,
response_model=KlingVideoExtendResponse,
),
request=KlingVideoExtendRequest(
prompt=prompt if prompt else None,
negative_prompt=negative_prompt if negative_prompt else None,
cfg_scale=cfg_scale,
video_id=video_id,
),
auth_kwargs=kwargs,
)
task_creation_response = initial_operation.execute()
validate_task_creation_response(task_creation_response)
task_id = task_creation_response.data.task_id
final_response = self.get_response(
task_id, auth_kwargs=kwargs, node_id=unique_id
)
validate_video_result_response(final_response)
video = get_video_from_response(final_response)
return video_result_to_node_output(video)
class KlingVideoEffectsBase(KlingNodeBase):
"""Kling Video Effects Base"""
RETURN_TYPES = ("VIDEO", "STRING", "STRING")
RETURN_NAMES = ("VIDEO", "video_id", "duration")
def get_response(
self, task_id: str, auth_kwargs: dict[str, str], node_id: Optional[str] = None
) -> KlingVideoEffectsResponse:
return poll_until_finished(
auth_kwargs,
ApiEndpoint(
path=f"{PATH_VIDEO_EFFECTS}/{task_id}",
method=HttpMethod.GET,
request_model=EmptyRequest,
response_model=KlingVideoEffectsResponse,
),
result_url_extractor=get_video_url_from_response,
estimated_duration=AVERAGE_DURATION_VIDEO_EFFECTS,
node_id=node_id,
)
def api_call(
self,
dual_character: bool,
effect_scene: KlingDualCharacterEffectsScene | KlingSingleImageEffectsScene,
model_name: str,
duration: KlingVideoGenDuration,
image_1: torch.Tensor,
image_2: Optional[torch.Tensor] = None,
mode: Optional[KlingVideoGenMode] = None,
unique_id: Optional[str] = None,
**kwargs,
):
if dual_character:
request_input_field = KlingDualCharacterEffectInput(
model_name=model_name,
mode=mode,
images=[
tensor_to_base64_string(image_1),
tensor_to_base64_string(image_2),
],
duration=duration,
)
else:
request_input_field = KlingSingleImageEffectInput(
model_name=model_name,
image=tensor_to_base64_string(image_1),
duration=duration,
)
initial_operation = SynchronousOperation(
endpoint=ApiEndpoint(
path=PATH_VIDEO_EFFECTS,
method=HttpMethod.POST,
request_model=KlingVideoEffectsRequest,
response_model=KlingVideoEffectsResponse,
),
request=KlingVideoEffectsRequest(
effect_scene=effect_scene,
input=request_input_field,
),
auth_kwargs=kwargs,
)
task_creation_response = initial_operation.execute()
validate_task_creation_response(task_creation_response)
task_id = task_creation_response.data.task_id
final_response = self.get_response(
task_id, auth_kwargs=kwargs, node_id=unique_id
)
validate_video_result_response(final_response)
video = get_video_from_response(final_response)
return video_result_to_node_output(video)
class KlingDualCharacterVideoEffectNode(KlingVideoEffectsBase):
"""Kling Dual Character Video Effect Node"""
@classmethod
def INPUT_TYPES(s):
return {
"required": {
"image_left": (IO.IMAGE, {"tooltip": "Left side image"}),
"image_right": (IO.IMAGE, {"tooltip": "Right side image"}),
"effect_scene": model_field_to_node_input(
IO.COMBO,
KlingVideoEffectsRequest,
"effect_scene",
enum_type=KlingDualCharacterEffectsScene,
),
"model_name": model_field_to_node_input(
IO.COMBO,
KlingDualCharacterEffectInput,
"model_name",
enum_type=KlingCharacterEffectModelName,
),
"mode": model_field_to_node_input(
IO.COMBO,
KlingDualCharacterEffectInput,
"mode",
enum_type=KlingVideoGenMode,
),
"duration": model_field_to_node_input(
IO.COMBO,
KlingDualCharacterEffectInput,
"duration",
enum_type=KlingVideoGenDuration,
),
},
"hidden": {
"auth_token": "AUTH_TOKEN_COMFY_ORG",
"comfy_api_key": "API_KEY_COMFY_ORG",
"unique_id": "UNIQUE_ID",
},
}
DESCRIPTION = "Achieve different special effects when generating a video based on the effect_scene. First image will be positioned on left side, second on right side of the composite."
RETURN_TYPES = ("VIDEO", "STRING")
RETURN_NAMES = ("VIDEO", "duration")
def api_call(
self,
image_left: torch.Tensor,
image_right: torch.Tensor,
effect_scene: KlingDualCharacterEffectsScene,
model_name: KlingCharacterEffectModelName,
mode: KlingVideoGenMode,
duration: KlingVideoGenDuration,
unique_id: Optional[str] = None,
**kwargs,
):
video, _, duration = super().api_call(
dual_character=True,
effect_scene=effect_scene,
model_name=model_name,
mode=mode,
duration=duration,
image_1=image_left,
image_2=image_right,
unique_id=unique_id,
**kwargs,
)
return video, duration
class KlingSingleImageVideoEffectNode(KlingVideoEffectsBase):
"""Kling Single Image Video Effect Node"""
@classmethod
def INPUT_TYPES(s):
return {
"required": {
"image": (
IO.IMAGE,
{
"tooltip": " Reference Image. URL or Base64 encoded string (without data:image prefix). File size cannot exceed 10MB, resolution not less than 300*300px, aspect ratio between 1:2.5 ~ 2.5:1"
},
),
"effect_scene": model_field_to_node_input(
IO.COMBO,
KlingVideoEffectsRequest,
"effect_scene",
enum_type=KlingSingleImageEffectsScene,
),
"model_name": model_field_to_node_input(
IO.COMBO,
KlingSingleImageEffectInput,
"model_name",
enum_type=KlingSingleImageEffectModelName,
),
"duration": model_field_to_node_input(
IO.COMBO,
KlingSingleImageEffectInput,
"duration",
enum_type=KlingVideoGenDuration,
),
},
"hidden": {
"auth_token": "AUTH_TOKEN_COMFY_ORG",
"comfy_api_key": "API_KEY_COMFY_ORG",
"unique_id": "UNIQUE_ID",
},
}
DESCRIPTION = "Achieve different special effects when generating a video based on the effect_scene."
def api_call(
self,
image: torch.Tensor,
effect_scene: KlingSingleImageEffectsScene,
model_name: KlingSingleImageEffectModelName,
duration: KlingVideoGenDuration,
unique_id: Optional[str] = None,
**kwargs,
):
return super().api_call(
dual_character=False,
effect_scene=effect_scene,
model_name=model_name,
duration=duration,
image_1=image,
unique_id=unique_id,
**kwargs,
)
class KlingLipSyncBase(KlingNodeBase):
"""Kling Lip Sync Base"""
RETURN_TYPES = ("VIDEO", "STRING", "STRING")
RETURN_NAMES = ("VIDEO", "video_id", "duration")
def validate_lip_sync_video(self, video: VideoInput):
"""
Validates the input video adheres to the expectations of the Kling Lip Sync API:
- Video length does not exceed 10s and is not shorter than 2s
- Length and width dimensions should both be between 720px and 1920px
See: https://app.klingai.com/global/dev/document-api/apiReference/model/videoTolip
"""
validate_video_dimensions(video, 720, 1920)
validate_video_duration(video, 2, 10)
def validate_text(self, text: str):
if not text:
raise ValueError("Text is required")
if len(text) > MAX_PROMPT_LENGTH_LIP_SYNC:
raise ValueError(
f"Text is too long. Maximum length is {MAX_PROMPT_LENGTH_LIP_SYNC} characters."
)
def get_response(
self, task_id: str, auth_kwargs: dict[str, str], node_id: Optional[str] = None
) -> KlingLipSyncResponse:
"""Polls the Kling API endpoint until the task reaches a terminal state."""
return poll_until_finished(
auth_kwargs,
ApiEndpoint(
path=f"{PATH_LIP_SYNC}/{task_id}",
method=HttpMethod.GET,
request_model=EmptyRequest,
response_model=KlingLipSyncResponse,
),
result_url_extractor=get_video_url_from_response,
estimated_duration=AVERAGE_DURATION_LIP_SYNC,
node_id=node_id,
)
def api_call(
self,
video: VideoInput,
audio: Optional[AudioInput] = None,
voice_language: Optional[str] = None,
mode: Optional[str] = None,
text: Optional[str] = None,
voice_speed: Optional[float] = None,
voice_id: Optional[str] = None,
unique_id: Optional[str] = None,
**kwargs,
) -> tuple[VideoFromFile, str, str]:
if text:
self.validate_text(text)
self.validate_lip_sync_video(video)
# Upload video to Comfy API and get download URL
video_url = upload_video_to_comfyapi(video, auth_kwargs=kwargs)
logging.info("Uploaded video to Comfy API. URL: %s", video_url)
# Upload the audio file to Comfy API and get download URL
if audio:
audio_url = upload_audio_to_comfyapi(audio, auth_kwargs=kwargs)
logging.info("Uploaded audio to Comfy API. URL: %s", audio_url)
else:
audio_url = None
initial_operation = SynchronousOperation(
endpoint=ApiEndpoint(
path=PATH_LIP_SYNC,
method=HttpMethod.POST,
request_model=KlingLipSyncRequest,
response_model=KlingLipSyncResponse,
),
request=KlingLipSyncRequest(
input=KlingLipSyncInputObject(
video_url=video_url,
mode=mode,
text=text,
voice_language=voice_language,
voice_speed=voice_speed,
audio_type="url",
audio_url=audio_url,
voice_id=voice_id,
),
),
auth_kwargs=kwargs,
)
task_creation_response = initial_operation.execute()
validate_task_creation_response(task_creation_response)
task_id = task_creation_response.data.task_id
final_response = self.get_response(
task_id, auth_kwargs=kwargs, node_id=unique_id
)
validate_video_result_response(final_response)
video = get_video_from_response(final_response)
return video_result_to_node_output(video)
class KlingLipSyncAudioToVideoNode(KlingLipSyncBase):
"""Kling Lip Sync Audio to Video Node. Syncs mouth movements in a video file to the audio content of an audio file."""
@classmethod
def INPUT_TYPES(s):
return {
"required": {
"video": (IO.VIDEO, {}),
"audio": (IO.AUDIO, {}),
"voice_language": model_field_to_node_input(
IO.COMBO,
KlingLipSyncInputObject,
"voice_language",
enum_type=KlingLipSyncVoiceLanguage,
),
},
"hidden": {
"auth_token": "AUTH_TOKEN_COMFY_ORG",
"comfy_api_key": "API_KEY_COMFY_ORG",
"unique_id": "UNIQUE_ID",
},
}
DESCRIPTION = "Kling Lip Sync Audio to Video Node. Syncs mouth movements in a video file to the audio content of an audio file. When using, ensure that the audio contains clearly distinguishable vocals and that the video contains a distinct face. The audio file should not be larger than 5MB. The video file should not be larger than 100MB, should have height/width between 720px and 1920px, and should be between 2s and 10s in length."
def api_call(
self,
video: VideoInput,
audio: AudioInput,
voice_language: str,
unique_id: Optional[str] = None,
**kwargs,
):
return super().api_call(
video=video,
audio=audio,
voice_language=voice_language,
mode="audio2video",
unique_id=unique_id,
**kwargs,
)
class KlingLipSyncTextToVideoNode(KlingLipSyncBase):
"""Kling Lip Sync Text to Video Node. Syncs mouth movements in a video file to a text prompt."""
@staticmethod
def get_voice_config() -> dict[str, tuple[str, str]]:
return {
# English voices
"Melody": ("girlfriend_4_speech02", "en"),
"Sunny": ("genshin_vindi2", "en"),
"Sage": ("zhinen_xuesheng", "en"),
"Ace": ("AOT", "en"),
"Blossom": ("ai_shatang", "en"),
"Peppy": ("genshin_klee2", "en"),
"Dove": ("genshin_kirara", "en"),
"Shine": ("ai_kaiya", "en"),
"Anchor": ("oversea_male1", "en"),
"Lyric": ("ai_chenjiahao_712", "en"),
"Tender": ("chat1_female_new-3", "en"),
"Siren": ("chat_0407_5-1", "en"),
"Zippy": ("cartoon-boy-07", "en"),
"Bud": ("uk_boy1", "en"),
"Sprite": ("cartoon-girl-01", "en"),
"Candy": ("PeppaPig_platform", "en"),
"Beacon": ("ai_huangzhong_712", "en"),
"Rock": ("ai_huangyaoshi_712", "en"),
"Titan": ("ai_laoguowang_712", "en"),
"Grace": ("chengshu_jiejie", "en"),
"Helen": ("you_pingjing", "en"),
"Lore": ("calm_story1", "en"),
"Crag": ("uk_man2", "en"),
"Prattle": ("laopopo_speech02", "en"),
"Hearth": ("heainainai_speech02", "en"),
"The Reader": ("reader_en_m-v1", "en"),
"Commercial Lady": ("commercial_lady_en_f-v1", "en"),
# Chinese voices
"阳光少年": ("genshin_vindi2", "zh"),
"懂事小弟": ("zhinen_xuesheng", "zh"),
"运动少年": ("tiyuxi_xuedi", "zh"),
"青春少女": ("ai_shatang", "zh"),
"温柔小妹": ("genshin_klee2", "zh"),
"元气少女": ("genshin_kirara", "zh"),
"阳光男生": ("ai_kaiya", "zh"),
"幽默小哥": ("tiexin_nanyou", "zh"),
"文艺小哥": ("ai_chenjiahao_712", "zh"),
"甜美邻家": ("girlfriend_1_speech02", "zh"),
"温柔姐姐": ("chat1_female_new-3", "zh"),
"职场女青": ("girlfriend_2_speech02", "zh"),
"活泼男童": ("cartoon-boy-07", "zh"),
"俏皮女童": ("cartoon-girl-01", "zh"),
"稳重老爸": ("ai_huangyaoshi_712", "zh"),
"温柔妈妈": ("you_pingjing", "zh"),
"严肃上司": ("ai_laoguowang_712", "zh"),
"优雅贵妇": ("chengshu_jiejie", "zh"),
"慈祥爷爷": ("zhuxi_speech02", "zh"),
"唠叨爷爷": ("uk_oldman3", "zh"),
"唠叨奶奶": ("laopopo_speech02", "zh"),
"和蔼奶奶": ("heainainai_speech02", "zh"),
"东北老铁": ("dongbeilaotie_speech02", "zh"),
"重庆小伙": ("chongqingxiaohuo_speech02", "zh"),
"四川妹子": ("chuanmeizi_speech02", "zh"),
"潮汕大叔": ("chaoshandashu_speech02", "zh"),
"台湾男生": ("ai_taiwan_man2_speech02", "zh"),
"西安掌柜": ("xianzhanggui_speech02", "zh"),
"天津姐姐": ("tianjinjiejie_speech02", "zh"),
"新闻播报男": ("diyinnansang_DB_CN_M_04-v2", "zh"),
"译制片男": ("yizhipiannan-v1", "zh"),
"撒娇女友": ("tianmeixuemei-v1", "zh"),
"刀片烟嗓": ("daopianyansang-v1", "zh"),
"乖巧正太": ("mengwa-v1", "zh"),
}
@classmethod
def INPUT_TYPES(s):
voice_options = list(s.get_voice_config().keys())
return {
"required": {
"video": (IO.VIDEO, {}),
"text": model_field_to_node_input(
IO.STRING, KlingLipSyncInputObject, "text", multiline=True
),
"voice": (voice_options, {"default": voice_options[0]}),
"voice_speed": model_field_to_node_input(
IO.FLOAT, KlingLipSyncInputObject, "voice_speed", slider=True
),
},
"hidden": {
"auth_token": "AUTH_TOKEN_COMFY_ORG",
"comfy_api_key": "API_KEY_COMFY_ORG",
"unique_id": "UNIQUE_ID",
},
}
DESCRIPTION = "Kling Lip Sync Text to Video Node. Syncs mouth movements in a video file to a text prompt. The video file should not be larger than 100MB, should have height/width between 720px and 1920px, and should be between 2s and 10s in length."
def api_call(
self,
video: VideoInput,
text: str,
voice: str,
voice_speed: float,
unique_id: Optional[str] = None,
**kwargs,
):
voice_id, voice_language = KlingLipSyncTextToVideoNode.get_voice_config()[voice]
return super().api_call(
video=video,
text=text,
voice_language=voice_language,
voice_id=voice_id,
voice_speed=voice_speed,
mode="text2video",
unique_id=unique_id,
**kwargs,
)
class KlingImageGenerationBase(KlingNodeBase):
"""Kling Image Generation Base Node."""
RETURN_TYPES = ("IMAGE",)
CATEGORY = "api node/image/Kling"
def validate_prompt(self, prompt: str, negative_prompt: Optional[str] = None):
if not prompt or len(prompt) > MAX_PROMPT_LENGTH_IMAGE_GEN:
raise ValueError(
f"Prompt must be less than {MAX_PROMPT_LENGTH_IMAGE_GEN} characters"
)
if negative_prompt and len(negative_prompt) > MAX_PROMPT_LENGTH_IMAGE_GEN:
raise ValueError(
f"Negative prompt must be less than {MAX_PROMPT_LENGTH_IMAGE_GEN} characters"
)
class KlingVirtualTryOnNode(KlingImageGenerationBase):
"""Kling Virtual Try On Node."""
@classmethod
def INPUT_TYPES(s):
return {
"required": {
"human_image": (IO.IMAGE, {}),
"cloth_image": (IO.IMAGE, {}),
"model_name": model_field_to_node_input(
IO.COMBO,
KlingVirtualTryOnRequest,
"model_name",
enum_type=KlingVirtualTryOnModelName,
),
},
"hidden": {
"auth_token": "AUTH_TOKEN_COMFY_ORG",
"comfy_api_key": "API_KEY_COMFY_ORG",
"unique_id": "UNIQUE_ID",
},
}
DESCRIPTION = "Kling Virtual Try On Node. Input a human image and a cloth image to try on the cloth on the human. You can merge multiple clothing item pictures into one image with a white background."
def get_response(
self, task_id: str, auth_kwargs: dict[str, str], node_id: Optional[str] = None
) -> KlingVirtualTryOnResponse:
return poll_until_finished(
auth_kwargs,
ApiEndpoint(
path=f"{PATH_VIRTUAL_TRY_ON}/{task_id}",
method=HttpMethod.GET,
request_model=EmptyRequest,
response_model=KlingVirtualTryOnResponse,
),
result_url_extractor=get_images_urls_from_response,
estimated_duration=AVERAGE_DURATION_VIRTUAL_TRY_ON,
node_id=node_id,
)
def api_call(
self,
human_image: torch.Tensor,
cloth_image: torch.Tensor,
model_name: KlingVirtualTryOnModelName,
unique_id: Optional[str] = None,
**kwargs,
):
initial_operation = SynchronousOperation(
endpoint=ApiEndpoint(
path=PATH_VIRTUAL_TRY_ON,
method=HttpMethod.POST,
request_model=KlingVirtualTryOnRequest,
response_model=KlingVirtualTryOnResponse,
),
request=KlingVirtualTryOnRequest(
human_image=tensor_to_base64_string(human_image),
cloth_image=tensor_to_base64_string(cloth_image),
model_name=model_name,
),
auth_kwargs=kwargs,
)
task_creation_response = initial_operation.execute()
validate_task_creation_response(task_creation_response)
task_id = task_creation_response.data.task_id
final_response = self.get_response(
task_id, auth_kwargs=kwargs, node_id=unique_id
)
validate_image_result_response(final_response)
images = get_images_from_response(final_response)
return (image_result_to_node_output(images),)
class KlingImageGenerationNode(KlingImageGenerationBase):
"""Kling Image Generation Node. Generate an image from a text prompt with an optional reference image."""
@classmethod
def INPUT_TYPES(s):
return {
"required": {
"prompt": model_field_to_node_input(
IO.STRING,
KlingImageGenerationsRequest,
"prompt",
multiline=True,
max_length=MAX_PROMPT_LENGTH_IMAGE_GEN,
),
"negative_prompt": model_field_to_node_input(
IO.STRING,
KlingImageGenerationsRequest,
"negative_prompt",
multiline=True,
),
"image_type": model_field_to_node_input(
IO.COMBO,
KlingImageGenerationsRequest,
"image_reference",
enum_type=KlingImageGenImageReferenceType,
),
"image_fidelity": model_field_to_node_input(
IO.FLOAT,
KlingImageGenerationsRequest,
"image_fidelity",
slider=True,
step=0.01,
),
"human_fidelity": model_field_to_node_input(
IO.FLOAT,
KlingImageGenerationsRequest,
"human_fidelity",
slider=True,
step=0.01,
),
"model_name": model_field_to_node_input(
IO.COMBO,
KlingImageGenerationsRequest,
"model_name",
enum_type=KlingImageGenModelName,
),
"aspect_ratio": model_field_to_node_input(
IO.COMBO,
KlingImageGenerationsRequest,
"aspect_ratio",
enum_type=KlingImageGenAspectRatio,
),
"n": model_field_to_node_input(
IO.INT,
KlingImageGenerationsRequest,
"n",
),
},
"optional": {
"image": (IO.IMAGE, {}),
},
"hidden": {
"auth_token": "AUTH_TOKEN_COMFY_ORG",
"comfy_api_key": "API_KEY_COMFY_ORG",
"unique_id": "UNIQUE_ID",
},
}
DESCRIPTION = "Kling Image Generation Node. Generate an image from a text prompt with an optional reference image."
def get_response(
self,
task_id: str,
auth_kwargs: Optional[dict[str, str]],
node_id: Optional[str] = None,
) -> KlingImageGenerationsResponse:
return poll_until_finished(
auth_kwargs,
ApiEndpoint(
path=f"{PATH_IMAGE_GENERATIONS}/{task_id}",
method=HttpMethod.GET,
request_model=EmptyRequest,
response_model=KlingImageGenerationsResponse,
),
result_url_extractor=get_images_urls_from_response,
estimated_duration=AVERAGE_DURATION_IMAGE_GEN,
node_id=node_id,
)
def api_call(
self,
model_name: KlingImageGenModelName,
prompt: str,
negative_prompt: str,
image_type: KlingImageGenImageReferenceType,
image_fidelity: float,
human_fidelity: float,
n: int,
aspect_ratio: KlingImageGenAspectRatio,
image: Optional[torch.Tensor] = None,
unique_id: Optional[str] = None,
**kwargs,
):
self.validate_prompt(prompt, negative_prompt)
if image is not None:
image = tensor_to_base64_string(image)
initial_operation = SynchronousOperation(
endpoint=ApiEndpoint(
path=PATH_IMAGE_GENERATIONS,
method=HttpMethod.POST,
request_model=KlingImageGenerationsRequest,
response_model=KlingImageGenerationsResponse,
),
request=KlingImageGenerationsRequest(
model_name=model_name,
prompt=prompt,
negative_prompt=negative_prompt,
image=image,
image_reference=image_type,
image_fidelity=image_fidelity,
human_fidelity=human_fidelity,
n=n,
aspect_ratio=aspect_ratio,
),
auth_kwargs=kwargs,
)
task_creation_response = initial_operation.execute()
validate_task_creation_response(task_creation_response)
task_id = task_creation_response.data.task_id
final_response = self.get_response(
task_id, auth_kwargs=kwargs, node_id=unique_id
)
validate_image_result_response(final_response)
images = get_images_from_response(final_response)
return (image_result_to_node_output(images),)
NODE_CLASS_MAPPINGS = {
"KlingCameraControls": KlingCameraControls,
"KlingTextToVideoNode": KlingTextToVideoNode,
"KlingImage2VideoNode": KlingImage2VideoNode,
"KlingCameraControlI2VNode": KlingCameraControlI2VNode,
"KlingCameraControlT2VNode": KlingCameraControlT2VNode,
"KlingStartEndFrameNode": KlingStartEndFrameNode,
"KlingVideoExtendNode": KlingVideoExtendNode,
"KlingLipSyncAudioToVideoNode": KlingLipSyncAudioToVideoNode,
"KlingLipSyncTextToVideoNode": KlingLipSyncTextToVideoNode,
"KlingVirtualTryOnNode": KlingVirtualTryOnNode,
"KlingImageGenerationNode": KlingImageGenerationNode,
"KlingSingleImageVideoEffectNode": KlingSingleImageVideoEffectNode,
"KlingDualCharacterVideoEffectNode": KlingDualCharacterVideoEffectNode,
}
NODE_DISPLAY_NAME_MAPPINGS = {
"KlingCameraControls": "Kling Camera Controls",
"KlingTextToVideoNode": "Kling Text to Video",
"KlingImage2VideoNode": "Kling Image to Video",
"KlingCameraControlI2VNode": "Kling Image to Video (Camera Control)",
"KlingCameraControlT2VNode": "Kling Text to Video (Camera Control)",
"KlingStartEndFrameNode": "Kling Start-End Frame to Video",
"KlingVideoExtendNode": "Kling Video Extend",
"KlingLipSyncAudioToVideoNode": "Kling Lip Sync Video with Audio",
"KlingLipSyncTextToVideoNode": "Kling Lip Sync Video with Text",
"KlingVirtualTryOnNode": "Kling Virtual Try On",
"KlingImageGenerationNode": "Kling Image Generation",
"KlingSingleImageVideoEffectNode": "Kling Video Effects",
"KlingDualCharacterVideoEffectNode": "Kling Dual Character Video Effects",
}
|