config.json

{
  "architectures": [
    "QianfanOCRForConditionalGeneration"
  ],
  "downsample_ratio": 0.5,
  "dynamic_image_size": true,
  "eos_token_id": 502,
  "force_image_size": 448,
  "image_token_id": 151671,
  "text_config": {
    "_name_or_path": "/root/codespace/checkpoints/Qwen3-4B",
    "architectures": [
      "Qwen3ForCausalLM"
    ],
    "attention_bias": false,
    "attention_dropout": 0.0,
    "bos_token_id": 151643,
    "debug": false,
    "eos_token_id": 151645,
    "ep_size": 1,
    "head_dim": 128,
    "hidden_act": "silu",
    "hidden_size": 2560,
    "initializer_range": 0.02,
    "intermediate_size": 9728,
    "max_position_embeddings": 32768,
    "max_window_layers": 36,
    "micro_forward": false,
    "model_type": "qwen3",
    "num_attention_heads": 32,
    "num_hidden_layers": 36,
    "num_key_value_heads": 8,
    "rms_norm_eps": 1e-06,
    "rope_scaling": null,
    "rope_theta": 5000000,
    "skip_checkpoint": false,
    "sliding_window": null,
    "torch_dtype": "bfloat16",
    "use_cache": false,
    "use_deepep": false,
    "use_sliding_window": false,
    "vocab_size": 153678
  },
  "max_dynamic_patch": 12,
  "min_dynamic_patch": 1,
  "model_type": "qianfan_ocr",
  "pad2square": false,
  "pad_token_id": 500,
  "ps_version": "v2",
  "select_layer": -1,
  "template": "internvl2_5",
  "tie_word_embeddings": false,
  "torch_dtype": "bfloat16",
  "transformers_version": null,
  "use_thumbnail": true,
  "vision_config": {
    "architectures": [
      "QianfanOCRVisionModel"
    ],
    "attention_dropout": 0.0,
    "drop_path_rate": 0.1,
    "hidden_act": "gelu",
    "hidden_size": 1024,
    "image_size": 448,
    "initializer_range": 0.02,
    "intermediate_size": 4096,
    "layer_norm_eps": 1e-06,
    "model_type": "qianfan_ocr_vision",
    "norm_type": "layer_norm",
    "num_attention_heads": 16,
    "num_channels": 3,
    "num_hidden_layers": 24,
    "patch_size": 14,
    "qkv_bias": true,
    "torch_dtype": "bfloat16"
  },
  "vocab_size": 500
}