{
  "architectures": ["Qwen3_5MoeForConditionalGeneration"],
  "image_token_id": 248056,
  "model_type": "qwen3_5_moe",
  "text_config": {
    "attention_bias": false,
    "attention_dropout": 0.0,
    "attn_output_gate": true,
    "bos_token_id": 248044,
    "dtype": "bfloat16",
    "eos_token_id": 248044,
    "full_attention_interval": 4,
    "head_dim": 256,
    "hidden_act": "silu",
    "hidden_size": 2048,
    "initializer_range": 0.02,
    "layer_types": [
      "linear_attention", "linear_attention", "linear_attention", "full_attention",
      "linear_attention", "linear_attention", "linear_attention", "full_attention",
      "linear_attention", "linear_attention", "linear_attention", "full_attention",
      "linear_attention", "linear_attention", "linear_attention", "full_attention",
      "linear_attention", "linear_attention", "linear_attention", "full_attention",
      "linear_attention", "linear_attention", "linear_attention", "full_attention",
      "linear_attention", "linear_attention", "linear_attention", "full_attention",
      "linear_attention", "linear_attention", "linear_attention", "full_attention",
      "linear_attention", "linear_attention", "linear_attention", "full_attention",
      "linear_attention", "linear_attention", "linear_attention", "full_attention"
    ],
    "linear_conv_kernel_dim": 4,
    "linear_key_head_dim": 128,
    "linear_num_key_heads": 16,
    "linear_num_value_heads": 32,
    "linear_value_head_dim": 128,
    "mamba_ssm_dtype": "float32",
    "max_position_embeddings": 262144,
    "model_type": "qwen3_5_moe_text",
    "moe_intermediate_size": 512,
    "mtp_num_hidden_layers": 1,
    "mtp_use_dedicated_embeddings": false,
    "num_attention_heads": 16,
    "num_experts": 256,
    "num_experts_per_tok": 8,
    "num_hidden_layers": 40,
    "num_key_value_heads": 2,
    "output_router_logits": false,
    "pad_token_id": null,
    "partial_rotary_factor": 0.25,
    "rms_norm_eps": 1e-06,
    "rope_parameters": {
      "mrope_interleaved": true,
      "mrope_section": [11, 11, 10],
      "partial_rotary_factor": 0.25,
      "rope_theta": 10000000,
      "rope_type": "default"
    },
    "router_aux_loss_coef": 0.001,
    "shared_expert_intermediate_size": 512,
    "tie_word_embeddings": false,
    "use_cache": true,
    "vocab_size": 248320
  },
  "tie_word_embeddings": false,
  "transformers_version": "4.57.1",
  "video_token_id": 248057,
  "vision_config": {
    "deepstack_visual_indexes": [],
    "depth": 27,
    "hidden_act": "gelu_pytorch_tanh",
    "hidden_size": 1152,
    "in_channels": 3,
    "initializer_range": 0.02,
    "intermediate_size": 4304,
    "model_type": "qwen3_5_moe",
    "num_heads": 16,
    "num_position_embeddings": 2304,
    "out_hidden_size": 2048,
    "patch_size": 16,
    "spatial_merge_size": 2,
    "temporal_patch_size": 2
  },
  "vision_end_token_id": 248054,
  "vision_start_token_id": 248053,
  "quantization_config": {
    "activation_scheme": "dynamic",
    "fmt": "e4m3",
    "quant_method": "fp8",
    "modules_to_not_convert": [
      "model.visual.blocks.0.attn.proj", "model.visual.blocks.0.attn.qkv",
      "model.visual.blocks.0.mlp.linear_fc1", "model.visual.blocks.0.mlp.linear_fc2",
      "visual.blocks.0.attn.proj", "visual.blocks.0.attn.qkv_proj",
      "visual.blocks.0.mlp.linear_fc1", "visual.blocks.0.mlp.linear_fc2",
      "model.visual.blocks.1.attn.proj", "model.visual.blocks.1.attn.qkv",
      "model.visual.blocks.1.mlp.linear_fc1", "model.visual.blocks.1.mlp.linear_fc2",
      "visual.blocks.1.attn.proj", "visual.blocks.1.attn.qkv_proj",
      "visual.blocks.1.mlp.linear_fc1", "visual.blocks.1.mlp.linear_fc2",
      ... (блоки 2..26 — по тому же шаблону: model.visual.blocks.N.attn.proj/qkv, .mlp.linear_fc1/fc2, visual.blocks.N.attn.proj/qkv_proj, .mlp.linear_fc1/fc2),
      "model.visual.deepstack_merger_list.0.linear_fc1",
      "model.visual.deepstack_merger_list.0.linear_fc2",
      "model.visual.deepstack_merger_list.0.norm",
      "visual.deepstack_merger_list.0.linear_fc1",
      "visual.deepstack_merger_list.0.linear_fc2",
      "visual.deepstack_merger_list.0.norm",
      "model.visual.deepstack_merger_list.1.linear_fc1",
      "model.visual.deepstack_merger_list.1.linear_fc2",
      "model.visual.deepstack_merger_list.1.norm",
      "visual.deepstack_merger_list.1.linear_fc1",
      "visual.deepstack_merger_list.1.linear_fc2",
      "visual.deepstack_merger_list.1.norm",
      "model.visual.deepstack_merger_list.2.linear_fc1",
      "model.visual.deepstack_merger_list.2.linear_fc2",
      "model.visual.deepstack_merger_list.2.norm",
      "visual.deepstack_merger_list.2.linear_fc1",
      "visual.deepstack_merger_list.2.linear_fc2",
      "visual.deepstack_merger_list.2.norm",
      "model.visual.merger.linear_fc1", "model.visual.merger.linear_fc2", "model.visual.merger.norm",
      "model.visual.patch_embed.proj", "model.visual.pos_embed",
      "visual.merger.linear_fc1", "visual.merger.linear_fc2", "visual.merger.norm",
      "visual.patch_embed.proj", "visual.pos_embed",
      "visual", "model.visual", "lm_head", "model.embed_tokens",
      "model.language_model.layers.0.input_layernorm",
      "model.language_model.layers.0.mlp.shared_expert_gate",
      "model.language_model.layers.0.post_attention_layernorm",
      "model.language_model.layers.0.mlp.gate",
      "model.language_model.layers.0.linear_attn.A_log",
      "model.language_model.layers.0.linear_attn.conv1d",
      "model.language_model.layers.0.linear_attn.dt_bias",
      "model.language_model.layers.0.linear_attn.in_proj_ba",
      "model.language_model.layers.0.linear_attn.in_proj_b",
      "model.language_model.layers.0.linear_attn.in_proj_a",
      "model.language_model.layers.0.linear_attn.norm",
      "model.language_model.layers.1.input_layernorm",
      "model.language_model.layers.1.mlp.shared_expert_gate",
      "model.language_model.layers.1.post_attention_layernorm",
      "model.language_model.layers.1.mlp.gate",
      "model.language_model.layers.1.linear_attn.A_log",
      "model.language_model.layers.1.linear_attn.conv1d",
      "model.language_model.layers.1.linear_attn.dt_bias",
      "model.language_model.layers.1.linear_attn.in_proj_ba",
      "model.language_model.layers.1.linear_attn.in_proj_b",
      "model.language_model.layers.1.linear_attn.in_proj_a",
      "model.language_model.layers.1.linear_attn.norm",
      ... (слои 2..39 — по тому же шаблону; на слоях-«full_attention» добавляются "self_attn.k_norm" и "self_attn.q_norm"),
      "model.language_model.layers.39.self_attn.q_norm",
      "mtp.layers.0.input_layernorm", "mtp.layers.0.mlp.gate",
      "mtp.layers.0.mlp.shared_expert_gate", "mtp.layers.0.post_attention_layernorm",
      "mtp.layers.0.self_attn.k_norm", "mtp.layers.0.self_attn.q_norm",
      "mtp.fc", "mtp.norm", "mtp.pre_fc_norm_embedding", "mtp.pre_fc_norm_hidden"
    ],
    "weight_block_size": [128, 128]
  }
}