Inconsistency between the pretrained models and labels

is there any code/tutorials to show how to use the pretrained pytorch models and test the evaluation performance for image classification on Imagenet and cifar10 100? (just testing not training, since I dont have the hardware for training)
For imagenet I am facing an inconsistency issue and I cannot find the details for the labels that are used to train the pytorch models.
In other words:
I notice that for pretrain models that are provided the labels are not consistent.
For example, vgg16 class 1 is different from Resnet50 class 1.
Can you let us know where we can find the corresponding labels for each model?
For vgg i notice that the one that looks like this:

  "0": [
    "n01440764",
    "tench"
  ],
  "1": [
    "n01443537",
    "goldfish"
  ],
  "2": [
    "n01484850",
    "great_white_shark"
  ],
  "3": [
    "n01491361",
    "tiger_shark"
  ],
  "4": [
    "n01494475",
    "hammerhead"
  ],
  "5": [
    "n01496331",
    "electric_ray"
  ],
  "6": [
    "n01498041",
    "stingray"
  ],
  "7": [
    "n01514668",
    "cock"
  ],
  "8": [
    "n01514859",
    "hen"

works, but this one is not the one that we should use for resnets, please let us know what we should do.
Thanks