// config.json
{
  "_name_or_path": "openai/clip-vit-base-patch32",
  "architectures": [
    "CLIPModel"
  ],
  "initializer_factor": 1.0,
  "model_type": "clip",
  "text_config": {
    "model_type": "clip_text_model",
    "initializer_factor": 1.0,
    "vocab_size": 49408,
    "max_position_embeddings": 77,
    "hidden_size": 512,
    "num_hidden_layers": 12,
    "num_attention_heads": 8,
    "intermediate_size": 2048,
    "projection_dim": 512
  },
  "vision_config": {
    "model_type": "clip_vision_model",
    "initializer_factor": 1.0,
    "image_size": 224,
    "patch_size": 32,
    "num_hidden_layers": 12,
    "num_attention_heads": 8,
    "hidden_size": 768,
    "intermediate_size": 3072,
    "projection_dim": 512
  }
}