{
  "architectures": [
    "VisionPixtralEncoderDecoder"
  ],
  "decoder": {
    "activation_dropout": 0.0,
    "activation_function": "relu",
    "add_cross_attention": true,
    "attention_dropout": 0.0,
    "classifier_dropout": 0.0,
    "cross_attention_hidden_size": 768,
    "d_model": 1024,
    "decoder_attention_heads": 16,
    "decoder_ffn_dim": 4096,
    "decoder_layerdrop": 0.0,
    "decoder_layers": 12,
    "dropout": 0.1,
    "init_std": 0.02,
    "is_decoder": true,
    "decoder_start_token_id": 0,
    "layernorm_embedding": false,
    "max_position_embeddings": 1024,
    "model_type": "trocr",
    "scale_embedding": true,
    "tie_word_embeddings": false,
    "torch_dtype": "float32",
    "use_cache": false,
    "use_learned_position_embeddings": false,
    "vocab_size": 50265
  },
  "encoder": {
    "_attn_implementation_autoset": true,
    "_name_or_path": "pixtral_encoder_decoder",
    "architectures": [
      "PixtralVisionModelBatch"
    ],
    "auto_map": {
      "AutoConfig": "configuration.PixtralVisionModelBatchConfig",
      "AutoModel": "modeling.PixtralVisionModelBatch"
    },
    "attention_dropout": 0.0,
    "head_dim": 64,
    "hidden_act": "silu",
    "hidden_size": 1024,
    "image_size": 1024,
    "initializer_range": 0.02,
    "intermediate_size": 4096,
    "is_composition": true,
    "model_type": "pixtral_batch",
    "num_attention_heads": 16,
    "num_channels": 3,
    "num_hidden_layers": 24,
    "patch_size": 16,
    "rope_theta": 10000.0,
    "tie_word_embeddings": false,
    "torch_dtype": "float32"
  },
  "is_encoder_decoder": true,
  "model_type": "vision_pixtral_encoder_decoder",
  "tie_word_embeddings": false,
  "torch_dtype": "float32",
  "transformers_version": "4.51.3"
}