{
  "architectures": [
    "RadioMBartPrefixForCausalLM"
  ],
  "auto_map": {
    "AutoConfig": "hf_nemotron_parse_config.NemotronParseConfig",
    "AutoModel": "hf_nemotron_parse_modeling.NemotronParseForConditionalGeneration",
    "AutoImageProcessor": "hf_nemotron_parse_processor.NemotronParseImageProcessor",
    "AutoProcessor": "hf_nemotron_parse_processor.NemotronParseProcessor"
  },
  "bos_token_id": 0,
  "decoder": {
    "_attn_implementation": "sdpa",
    "_name_or_path": "",
    "activation_dropout": 0.0,
    "activation_function": "gelu",
    "add_cross_attention": false,
    "add_final_layer_norm": true,
    "architectures": null,
    "attention_dropout": 0.0,
    "bad_words_ids": null,
    "begin_suppress_tokens": null,
    "bos_token_id": 0,
    "chunk_size_feed_forward": 0,
    "classifier_dropout": 0.0,
    "cross_attention_hidden_size": null,
    "d_model": 1024,
    "decoder_attention_heads": 16,
    "decoder_ffn_dim": 4096,
    "decoder_layerdrop": 0.0,
    "decoder_layers": 10,
    "decoder_start_token_id": null,
    "diversity_penalty": 0.0,
    "do_sample": false,
    "dropout": 0.1,
    "early_stopping": false,
    "encoder_attention_heads": 16,
    "encoder_ffn_dim": 4096,
    "encoder_layerdrop": 0.0,
    "encoder_layers": 12,
    "encoder_no_repeat_ngram_size": 0,
    "eos_token_id": 2,
    "exponential_decay_length_penalty": null,
    "finetuning_task": null,
    "forced_bos_token_id": null,
    "forced_eos_token_id": 2,
    "hidden_size": 1024,
    "id2label": {
      "0": "LABEL_0",
      "1": "LABEL_1",
      "2": "LABEL_2"
    },
    "init_std": 0.02,
    "is_decoder": true,
    "is_encoder_decoder": false,
    "label2id": {
      "LABEL_0": 0,
      "LABEL_1": 1,
      "LABEL_2": 2
    },
    "length_penalty": 1.0,
    "max_length": 20,
    "min_length": 0,
    "model_type": "nemotron_parse_text",
    "no_repeat_ngram_size": 0,
    "num_beam_groups": 1,
    "num_beams": 1,
    "num_hidden_layers": 10,
    "num_return_sequences": 1,
    "output_attentions": false,
    "output_hidden_states": false,
    "output_scores": false,
    "pad_token_id": 1,
    "prefix": null,
    "problem_type": null,
    "pruned_heads": {},
    "remove_invalid_values": false,
    "repetition_penalty": 1.0,
    "return_dict": true,
    "return_dict_in_generate": false,
    "scale_embedding": true,
    "sep_token_id": null,
    "suppress_tokens": null,
    "task_specific_params": null,
    "temperature": 1.0,
    "tf_legacy_loss": false,
    "tie_encoder_decoder": false,
    "tie_word_embeddings": true,
    "tokenizer_class": null,
    "top_k": 50,
    "top_p": 1.0,
    "torch_dtype": "bfloat16",
    "torchscript": false,
    "transformers_version": "5.6.1",
    "typical_p": 1.0,
    "use_bfloat16": true,
    "use_cache": true,
    "vocab_size": 72512,
    "self_attention_kv_heads": 8,
    "cross_attention_kv_heads": 0,
    "causal_vision_prefix": true
  },
  "decoder_start_token_id": 2,
  "encoder": {
    "_name_or_path": "nvidia/C-RADIOv4-SO400M",
    "adaptor_configs": {},
    "adaptor_names": null,
    "architectures": [
      "RADIOModel"
    ],
    "args": {
      "cls_token_per_teacher": true,
      "cpe_max_size": 2048,
      "model": "vit_so400m_patch16_224",
      "register_multiple": 10,
      "teachers": [
        {
          "name": "siglip2-g",
          "use_summary": true
        },
        {
          "name": "dino_v3_7b",
          "use_summary": true
        },
        {
          "name": "sam3",
          "use_summary": false
        }
      ]
    },
    "auto_map": {
      "AutoConfig": "nvidia/C-RADIOv4-SO400M--hf_model.RADIOConfig",
      "AutoModel": "nvidia/C-RADIOv4-SO400M--hf_model.RADIOModel"
    },
    "chunk_size_feed_forward": 0,
    "dtype": "float32",
    "feature_normalizer_config": null,
    "id2label": {
      "0": "LABEL_0",
      "1": "LABEL_1"
    },
    "inter_feature_normalizer_config": null,
    "is_encoder_decoder": false,
    "label2id": {
      "LABEL_0": 0,
      "LABEL_1": 1
    },
    "max_resolution": 2048,
    "model_type": "",
    "output_attentions": false,
    "output_hidden_states": false,
    "patch_size": 16,
    "preferred_resolution": [
      512,
      512
    ],
    "problem_type": null,
    "return_dict": true,
    "version": "c-radio_v4-so400m",
    "vitdet_window_size": null,
    "code_revision": "c0457f5dc26ca145f954cd4fc5bb6114e5705ad8"
  },
  "eos_token_id": 2,
  "image_size": [
    2048,
    1664
  ],
  "is_encoder_decoder": false,
  "max_sequence_length": 16384,
  "model_type": "nemotron_parse",
  "pad_token_id": 1,
  "tie_word_embeddings": true,
  "torch_dtype": "bfloat16",
  "transformers_version": "5.6.1",
  "vocab_size": 72512,
  "class_token_start_idx": 52315,
  "legacy_use_hf_vision_model": true,
  "vision_backbone_name": "nvidia/C-RADIOv4-SO400M",
  "vision_spatial_dim": 1152,
  "vision_summary_dim": 2304,
  "architecture_benchmark": {
    "backbone": "nvidia/C-RADIOv4-SO400M",
    "attention_mode": "global",
    "spatial_dim": 1152,
    "summary_dim": 2304
  },
  "navit_disable_cpe_augmentation": true,
  "self_attention_kv_heads": 8,
  "cross_attention_kv_heads": 0,
  "legacy_causal_vision_prefix": true,
  "legacy_causal_prefix_context_limit": 16384,
  "image_token_index": 1,
  "num_hidden_layers": 10,
  "decoder_layers": 10,
  "hidden_size": 1024,
  "intermediate_size": 4096,
  "num_attention_heads": 16,
  "num_key_value_heads": 8,
  "max_position_embeddings": 16384,
  "navit_position_embedding_mode": "content_normalized_2d_interpolate",
  "navit_packing": true,
  "navit_coordinate_frame": "content_normalized",
  "virtual_canvas_width": 1664,
  "virtual_canvas_height": 2048,
  "virtual_canvas_placement": "aligned_center",
  "num_extra_heads": 1
}