{
  "architectures": [
    "GroundAnythingForConditionalGeneration"
  ],
  "auto_map": {
    "AutoConfig": "configuration_groundinganything.GroundAnythingConfig",
    "AutoModel": "modeling_groundinganything.GroundAnythingModel",
    "AutoModelForCausalLM": "modeling_groundinganything.GroundAnythingForConditionalGeneration",
    "AutoModelForImageTextToText": "modeling_groundinganything.GroundAnythingForConditionalGeneration",
    "AutoProcessor": "processing_groundinganything.GroundAnythingProcessor"
  },
  "bos_token_id": null,
  "dtype": "bfloat16",
  "eos_token_id": 151645,
  "hidden_size": 2560,
  "image_token_id": 151655,
  "model_type": "groundinganything",
  "pad_token_id": 151643,
  "text_config": {
    "_name_or_path": "Qwen3-4B-Instruct-2507",
    "architectures": [
      "Qwen3ForCausalLM"
    ],
    "attention_bias": false,
    "attention_dropout": 0.0,
    "bos_token_id": 151643,
    "dtype": "bfloat16",
    "eos_token_id": 151645,
    "head_dim": 128,
    "hidden_act": "silu",
    "hidden_size": 2560,
    "initializer_range": 0.02,
    "intermediate_size": 9728,
    "layer_types": [
      "full_attention",
      "full_attention",
      "full_attention",
      "full_attention",
      "full_attention",
      "full_attention",
      "full_attention",
      "full_attention",
      "full_attention",
      "full_attention",
      "full_attention",
      "full_attention",
      "full_attention",
      "full_attention",
      "full_attention",
      "full_attention",
      "full_attention",
      "full_attention",
      "full_attention",
      "full_attention",
      "full_attention",
      "full_attention",
      "full_attention",
      "full_attention",
      "full_attention",
      "full_attention",
      "full_attention",
      "full_attention",
      "full_attention",
      "full_attention",
      "full_attention",
      "full_attention",
      "full_attention",
      "full_attention",
      "full_attention",
      "full_attention"
    ],
    "max_position_embeddings": 262144,
    "max_window_layers": 36,
    "model_type": "qwen3",
    "num_attention_heads": 32,
    "num_hidden_layers": 36,
    "num_key_value_heads": 8,
    "pad_token_id": 151643,
    "rms_norm_eps": 1e-06,
    "rope_parameters": {
      "rope_theta": 5000000,
      "rope_type": "default"
    },
    "sliding_window": null,
    "tie_word_embeddings": false,
    "use_cache": false,
    "use_sliding_window": false,
    "vocab_size": 152670
  },
  "tie_word_embeddings": false,
  "transformers_version": "5.7.0",
  "use_cache": false,
  "video_token_id": 151656,
  "vision_config": {
    "activation_func": "gelu_pytorch_tanh",
    "attention_dropout": 0.0,
    "attn_bias": false,
    "dtype": "bfloat16",
    "frame_windows_size": 4,
    "hidden_act": "gelu",
    "hidden_size": 1024,
    "image_size": 448,
    "init_pos_emb_height": 64,
    "init_pos_emb_time": 4,
    "init_pos_emb_width": 64,
    "initializer_range": 0.02,
    "intermediate_size": 4096,
    "layer_norm_eps": 1e-06,
    "layer_norm_type": "layer_norm",
    "linear_bias": false,
    "max_position_embeddings": 8192,
    "merge_kernel_size": [
      2,
      2
    ],
    "merge_type": "sd2_tpool",
    "mlp_type": "mlp2",
    "model_type": "groundinganything_vision",
    "norm_type": "rmsnorm",
    "num_attention_heads": 12,
    "num_channels": 3,
    "num_hidden_layers": 27,
    "out_hidden_size": 2560,
    "patch_embed_proj_bias": false,
    "patch_position_encoding_type": "absolute",
    "patch_size": 14,
    "pos_emb_interpolation_mode": "bilinear",
    "pos_emb_type": "divided_fixed",
    "projector_hidden_act": "gelu",
    "projector_hidden_size": 4096,
    "projector_ln_eps": 1e-05,
    "qkv_hidden_size": 1536,
    "rope_theta": 10000.0,
    "spatial_merge_size": 2,
    "tokens_per_second": 1,
    "use_head": false,
    "use_patch_position_encoding": false
  },
  "vision_end_token_id": 151653,
  "vision_start_token_id": 151652
}