config.json

{
  "architectures": [
    "Ovis2_5"
  ],
  "auto_map": {
    "AutoConfig": "configuration_ovis2_5.Ovis2_5_Config",
    "AutoModelForCausalLM": "modeling_ovis2_5.Ovis2_5"
  },
  "conversation_formatter_class": "Qwen3ConversationFormatter",
  "hidden_size": 2048,
  "vocab_size": 151936,
  "num_attention_heads": 32,
  "max_position_embeddings": 40960,
  "llm_config": {
    "_attn_implementation_autoset": true,
    "_name_or_path": "Qwen/Qwen3-1.7B",
    "architectures": [
      "Qwen3ForCausalLM"
    ],
    "attention_bias": false,
    "attention_dropout": 0.0,
    "bos_token_id": 151643,
    "eos_token_id": 151645,
    "head_dim": 128,
    "hidden_act": "silu",
    "hidden_size": 2048,
    "initializer_range": 0.02,
    "intermediate_size": 6144,
    "max_position_embeddings": 40960,
    "max_window_layers": 28,
    "model_type": "qwen3",
    "num_attention_heads": 16,
    "num_hidden_layers": 28,
    "num_key_value_heads": 8,
    "rms_norm_eps": 1e-06,
    "rope_scaling": null,
    "rope_theta": 1000000,
    "sliding_window": null,
    "tie_word_embeddings": true,
    "torch_dtype": "float32",
    "use_cache": true,
    "use_sliding_window": false,
    "vocab_size": 151936
  },
  "model_type": "ovis2_5",
  "torch_dtype": "bfloat16",
  "transformers_version": "4.51.3",
  "use_cache": true,
  "visual_vocab_size": 65536,
  "vit_config": {
    "_attn_implementation_autoset": true,
    "_name_or_path": "google/siglip2-so400m-patch16-512",
    "attention_dropout": 0.0,
    "fullatt_block_indexes": null,
    "hidden_act": "gelu_pytorch_tanh",
    "hidden_size": 1152,
    "hidden_stride": 2,
    "image_size": 512,
    "intermediate_size": 4304,
    "layer_norm_eps": 1e-06,
    "model_type": "siglip2_navit",
    "num_attention_heads": 16,
    "num_channels": 3,
    "num_hidden_layers": 27,
    "num_patches": -1,
    "patch_size": 16,
    "preserve_original_pe": true,
    "temporal_patch_size": 1,
    "torch_dtype": "float32",
    "use_rope": true,
    "window_size": 112
  }
}