{
  "architectures": [
    "Qwen3_5MoeForConditionalGeneration"
  ],
  "dtype": "bfloat16",
  "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": "5.7.0.dev0",
  "video_token_id": 248057,
  "vision_config": {
    "deepstack_visual_indexes": [],
    "depth": 27,
    "dtype": "bfloat16",
    "hidden_act": "gelu_pytorch_tanh",
    "hidden_size": 1152,
    "in_channels": 3,
    "initializer_range": 0.02,
    "intermediate_size": 4304,
    "model_type": "qwen3_5_moe_vision",
    "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": {
    "config_groups": {
      "group_0": {
        "input_activations": {
          "dynamic": false,
          "num_bits": 8,
          "type": "float"
        },
        "weights": {
          "dynamic": false,
          "num_bits": 8,
          "type": "float"
        },
        "targets": [
          "model.language_model.layers.0.linear_attn.in_proj_qkv",
          "model.language_model.layers.0.linear_attn.in_proj_z",
          "model.language_model.layers.0.linear_attn.out_proj",
          "model.language_model.layers.1.linear_attn.in_proj_qkv",
          "model.language_model.layers.1.linear_attn.in_proj_z",
          "model.language_model.layers.1.linear_attn.out_proj",
          "model.language_model.layers.10.linear_attn.in_proj_qkv",
          "model.language_model.layers.10.linear_attn.in_proj_z",
          "model.language_model.layers.10.linear_attn.out_proj",
          "model.language_model.layers.11.self_attn.k_proj",
          "model.language_model.layers.11.self_attn.o_proj",
          "model.language_model.layers.11.self_attn.q_proj",
          "model.language_model.layers.11.self_attn.v_proj",
          "model.language_model.layers.12.linear_attn.in_proj_qkv",
          "model.language_model.layers.12.linear_attn.in_proj_z",
          "model.language_model.layers.12.linear_attn.out_proj",
          "model.language_model.layers.13.linear_attn.in_proj_qkv",
          "model.language_model.layers.13.linear_attn.in_proj_z",
          "model.language_model.layers.13.linear_attn.out_proj",
          "model.language_model.layers.14.linear_attn.in_proj_qkv",
          "model.language_model.layers.14.linear_attn.in_proj_z",
          "model.language_model.layers.14.linear_attn.out_proj",
          "model.language_model.layers.15.self_attn.k_proj",
          "model.language_model.layers.15.self_attn.o_proj",
          "model.language_model.layers.15.self_attn.q_proj",
          "model.language_model.layers.15.self_attn.v_proj",
          "model.language_model.layers.16.linear_attn.in_proj_qkv",
          "model.language_model.layers.16.linear_attn.in_proj_z",
          "model.language_model.layers.16.linear_attn.out_proj",
          "model.language_model.layers.17.linear_attn.in_proj_qkv",
          "model.language_model.layers.17.linear_attn.in_proj_z",
          "model.language_model.layers.17.linear_attn.out_proj",
          "model.language_model.layers.18.linear_attn.in_proj_qkv",
          "model.language_model.layers.18.linear_attn.in_proj_z",
          "model.language_model.layers.18.linear_attn.out_proj",
          "model.language_model.layers.19.self_attn.k_proj",
          "model.language_model.layers.19.self_attn.o_proj",
          "model.language_model.layers.19.self_attn.q_proj",
          "model.language_model.layers.19.self_attn.v_proj",
          "model.language_model.layers.2.linear_attn.in_proj_qkv",
          "model.language_model.layers.2.linear_attn.in_proj_z",
          "model.language_model.layers.2.linear_attn.out_proj",
          "model.language_model.layers.20.linear_attn.in_proj_qkv",
          "model.language_model.layers.20.linear_attn.in_proj_z",
          "model.language_model.layers.20.linear_attn.out_proj",
          "model.language_model.layers.21.linear_attn.in_proj_qkv",
          "model.language_model.layers.21.linear_attn.in_proj_z",
          "model.language_model.layers.21.linear_attn.out_proj",
          "model.language_model.layers.22.linear_attn.in_proj_qkv",
          "model.language_model.layers.22.linear_attn.in_proj_z",
          "model.language_model.layers.22.linear_attn.out_proj",
          "model.language_model.layers.23.self_attn.k_proj",
          "model.language_model.layers.23.self_attn.o_proj",
          "model.language_model.layers.23.self_attn.q_proj",
          "model.language_model.layers.23.self_attn.v_proj",
          "model.language_model.layers.24.linear_attn.in_proj_qkv",
          "model.language_model.layers.24.linear_attn.in_proj_z",
          "model.language_model.layers.24.linear_attn.out_proj",
          "model.language_model.layers.25.linear_attn.in_proj_qkv",
          "model.language_model.layers.25.linear_attn.in_proj_z",
          "model.language_model.layers.25.linear_attn.out_proj",
          "model.language_model.layers.26.linear_attn.in_proj_qkv",
          "model.language_model.layers.26.linear_attn.in_proj_z",
          "model.language_model.layers.26.linear_attn.out_proj",
          "model.language_model.layers.27.self_attn.k_proj",
          "model.language_model.layers.27.self_attn.o_proj",
          "model.language_model.layers.27.self_attn.q_proj",
          "model.language_model.layers.27.self_attn.v_proj",
          "model.language_model.layers.28.linear_attn.in_proj_qkv",
          "model.language_model.layers.28.linear_attn.in_proj_z",
          "model.language_model.layers.28.linear_attn.out_proj",
          "model.language_model.layers.29.linear_attn.in_proj_qkv",
          "model.language_model.layers.29.linear_attn.in_proj_z",
          "model.language_model.layers.29.linear_attn.out_proj",
          "model.language_model.layers.3.self_attn.k_proj",
          "model.language_model.layers.3.self_attn.o_proj",
          "model.language_model.layers.3.self_attn.q_proj",
          "model.language_model.layers.3.self_attn.v_proj",
          "model.language_model.layers.30.linear_attn.in_proj_qkv",
          "model.language_model.layers.30.linear_attn.in_proj_z",
          "model.language_model.layers.30.linear_attn.out_proj",
          "model.language_model.layers.31.self_attn.k_proj",
          "model.language_model.layers.31.self_attn.o_proj",
          "model.language_model.layers.31.self_attn.q_proj",
          "model.language_model.layers.31.self_attn.v_proj",
          "model.language_model.layers.32.linear_attn.in_proj_qkv",
          "model.language_model.layers.32.linear_attn.in_proj_z",
          "model.language_model.layers.32.linear_attn.out_proj",
          "model.language_model.layers.33.linear_attn.in_proj_qkv",
          "model.language_model.layers.33.linear_attn.in_proj_z",
          "model.language_model.layers.33.linear_attn.out_proj",
          "model.language_model.layers.34.linear_attn.in_proj_qkv",
          "model.language_model.layers.34.linear_attn.in_proj_z",
          "model.language_model.layers.34.linear_attn.out_proj",
          "model.language_model.layers.35.self_attn.k_proj",
          "model.language_model.layers.35.self_attn.o_proj",
          "model.language_model.layers.35.self_attn.q_proj",
          "model.language_model.layers.35.self_attn.v_proj",
          "model.language_model.layers.36.linear_attn.in_proj_qkv",
          "model.language_model.layers.36.linear_attn.in_proj_z",
          "model.language_model.layers.36.linear_attn.out_proj",
          "model.language_model.layers.37.linear_attn.in_proj_qkv",
          "model.language_model.layers.37.linear_attn.in_proj_z",
          "model.language_model.layers.37.linear_attn.out_proj",
          "model.language_model.layers.38.linear_attn.in_proj_qkv",
          "model.language_model.layers.38.linear_attn.in_proj_z",
          "model.language_model.layers.38.linear_attn.out_proj",
          "model.language_model.layers.39.self_attn.k_proj",
          "model.language_model.layers.39.self_attn.o_proj",
          "model.language_model.layers.39.self_attn.q_proj",
          "model.language_model.layers.39.self_attn.v_proj",
          "model.language_model.layers.4.linear_attn.in_proj_qkv",
          "model.language_model.layers.4.linear_attn.in_proj_z",
          "model.language_model.layers.4.linear_attn.out_proj",
          "model.language_model.layers.5.linear_attn.in_proj_qkv",
          "model.language_model.layers.5.linear_attn.in_proj_z",
          "model.language_model.layers.5.linear_attn.out_proj",
          "model.language_model.layers.6.linear_attn.in_proj_qkv",
          "model.language_model.layers.6.linear_attn.in_proj_z",
          "model.language_model.layers.6.linear_attn.out_proj",
          "model.language_model.layers.7.self_attn.k_proj",
          "model.language_model.layers.7.self_attn.o_proj",
          "model.language_model.layers.7.self_attn.q_proj",
          "model.language_model.layers.7.self_attn.v_proj",
          "model.language_model.layers.8.linear_attn.in_proj_qkv",
          "model.language_model.layers.8.linear_attn.in_proj_z",
          "model.language_model.layers.8.linear_attn.out_proj",
          "model.language_model.layers.9.linear_attn.in_proj_qkv",
          "model.language_model.layers.9.linear_attn.in_proj_z",
          "model.language_model.layers.9.linear_attn.out_proj"
        ]
      },
      "group_1": {
        "input_activations": {
          "dynamic": false,
          "num_bits": 4,
          "type": "float",
          "group_size": 16
        },
        "weights": {
          "dynamic": false,
          "num_bits": 4,
          "type": "float",
          "group_size": 16
        },
        "targets": [
          "lm_head",
          "model.language_model.layers.0.mlp.experts",
          "model.language_model.layers.0.mlp.shared_expert.down_proj",
          "model.language_model.layers.0.mlp.shared_expert.gate_proj",
          "model.language_model.layers.0.mlp.shared_expert.up_proj",
          "model.language_model.layers.1.mlp.experts",
          "model.language_model.layers.1.mlp.shared_expert.down_proj",
          "model.language_model.layers.1.mlp.shared_expert.gate_proj",
          "model.language_model.layers.1.mlp.shared_expert.up_proj",
          "model.language_model.layers.10.mlp.experts",
          "model.language_model.layers.10.mlp.shared_expert.down_proj",
          "model.language_model.layers.10.mlp.shared_expert.gate_proj",
          "model.language_model.layers.10.mlp.shared_expert.up_proj",
          "model.language_model.layers.11.mlp.experts",
          "model.language_model.layers.11.mlp.shared_expert.down_proj",
          "model.language_model.layers.11.mlp.shared_expert.gate_proj",
          "model.language_model.layers.11.mlp.shared_expert.up_proj",
          "model.language_model.layers.12.mlp.experts",
          "model.language_model.layers.12.mlp.shared_expert.down_proj",
          "model.language_model.layers.12.mlp.shared_expert.gate_proj",
          "model.language_model.layers.12.mlp.shared_expert.up_proj",
          "model.language_model.layers.13.mlp.experts",
          "model.language_model.layers.13.mlp.shared_expert.down_proj",
          "model.language_model.layers.13.mlp.shared_expert.gate_proj",
          "model.language_model.layers.13.mlp.shared_expert.up_proj",
          "model.language_model.layers.14.mlp.experts",
          "model.language_model.layers.14.mlp.shared_expert.down_proj",
          "model.language_model.layers.14.mlp.shared_expert.gate_proj",
          "model.language_model.layers.14.mlp.shared_expert.up_proj",
          "model.language_model.layers.15.mlp.experts",
          "model.language_model.layers.15.mlp.shared_expert.down_proj",
          "model.language_model.layers.15.mlp.shared_expert.gate_proj",
          "model.language_model.layers.15.mlp.shared_expert.up_proj",
          "model.language_model.layers.16.mlp.experts",
          "model.language_model.layers.16.mlp.shared_expert.down_proj",
          "model.language_model.layers.16.mlp.shared_expert.gate_proj",
          "model.language_model.layers.16.mlp.shared_expert.up_proj",
          "model.language_model.layers.17.mlp.experts",
          "model.language_model.layers.17.mlp.shared_expert.down_proj",
          "model.language_model.layers.17.mlp.shared_expert.gate_proj",
          "model.language_model.layers.17.mlp.shared_expert.up_proj",
          "model.language_model.layers.18.mlp.experts",
          "model.language_model.layers.18.mlp.shared_expert.down_proj",
          "model.language_model.layers.18.mlp.shared_expert.gate_proj",
          "model.language_model.layers.18.mlp.shared_expert.up_proj",
          "model.language_model.layers.19.mlp.experts",
          "model.language_model.layers.19.mlp.shared_expert.down_proj",
          "model.language_model.layers.19.mlp.shared_expert.gate_proj",
          "model.language_model.layers.19.mlp.shared_expert.up_proj",
          "model.language_model.layers.2.mlp.experts",
          "model.language_model.layers.2.mlp.shared_expert.down_proj",
          "model.language_model.layers.2.mlp.shared_expert.gate_proj",
          "model.language_model.layers.2.mlp.shared_expert.up_proj",
          "model.language_model.layers.20.mlp.experts",
          "model.language_model.layers.20.mlp.shared_expert.down_proj",
          "model.language_model.layers.20.mlp.shared_expert.gate_proj",
          "model.language_model.layers.20.mlp.shared_expert.up_proj",
          "model.language_model.layers.21.mlp.experts",
          "model.language_model.layers.21.mlp.shared_expert.down_proj",
          "model.language_model.layers.21.mlp.shared_expert.gate_proj",
          "model.language_model.layers.21.mlp.shared_expert.up_proj",
          "model.language_model.layers.22.mlp.experts",
          "model.language_model.layers.22.mlp.shared_expert.down_proj",
          "model.language_model.layers.22.mlp.shared_expert.gate_proj",
          "model.language_model.layers.22.mlp.shared_expert.up_proj",
          "model.language_model.layers.23.mlp.experts",
          "model.language_model.layers.23.mlp.shared_expert.down_proj",
          "model.language_model.layers.23.mlp.shared_expert.gate_proj",
          "model.language_model.layers.23.mlp.shared_expert.up_proj",
          "model.language_model.layers.24.mlp.experts",
          "model.language_model.layers.24.mlp.shared_expert.down_proj",
          "model.language_model.layers.24.mlp.shared_expert.gate_proj",
          "model.language_model.layers.24.mlp.shared_expert.up_proj",
          "model.language_model.layers.25.mlp.experts",
          "model.language_model.layers.25.mlp.shared_expert.down_proj",
          "model.language_model.layers.25.mlp.shared_expert.gate_proj",
          "model.language_model.layers.25.mlp.shared_expert.up_proj",
          "model.language_model.layers.26.mlp.experts",
          "model.language_model.layers.26.mlp.shared_expert.down_proj",
          "model.language_model.layers.26.mlp.shared_expert.gate_proj",
          "model.language_model.layers.26.mlp.shared_expert.up_proj",
          "model.language_model.layers.27.mlp.experts",
          "model.language_model.layers.27.mlp.shared_expert.down_proj",
          "model.language_model.layers.27.mlp.shared_expert.gate_proj",
          "model.language_model.layers.27.mlp.shared_expert.up_proj",
          "model.language_model.layers.28.mlp.experts",
          "model.language_model.layers.28.mlp.shared_expert.down_proj",
          "model.language_model.layers.28.mlp.shared_expert.gate_proj",
          "model.language_model.layers.28.mlp.shared_expert.up_proj",
          "model.language_model.layers.29.mlp.experts",
          "model.language_model.layers.29.mlp.shared_expert.down_proj",
          "model.language_model.layers.29.mlp.shared_expert.gate_proj",
          "model.language_model.layers.29.mlp.shared_expert.up_proj",
          "model.language_model.layers.3.mlp.experts",
          "model.language_model.layers.3.mlp.shared_expert.down_proj",
          "model.language_model.layers.3.mlp.shared_expert.gate_proj",
          "model.language_model.layers.3.mlp.shared_expert.up_proj",
          "model.language_model.layers.30.mlp.experts",
          "model.language_model.layers.30.mlp.shared_expert.down_proj",
          "model.language_model.layers.30.mlp.shared_expert.gate_proj",
          "model.language_model.layers.30.mlp.shared_expert.up_proj",
          "model.language_model.layers.31.mlp.experts",
          "model.language_model.layers.31.mlp.shared_expert.down_proj",
          "model.language_model.layers.31.mlp.shared_expert.gate_proj",
          "model.language_model.layers.31.mlp.shared_expert.up_proj",
          "model.language_model.layers.32.mlp.experts",
          "model.language_model.layers.32.mlp.shared_expert.down_proj",
          "model.language_model.layers.32.mlp.shared_expert.gate_proj",
          "model.language_model.layers.32.mlp.shared_expert.up_proj",
          "model.language_model.layers.33.mlp.experts",
          "model.language_model.layers.33.mlp.shared_expert.down_proj",
          "model.language_model.layers.33.mlp.shared_expert.gate_proj",
          "model.language_model.layers.33.mlp.shared_expert.up_proj",
          "model.language_model.layers.34.mlp.experts",
          "model.language_model.layers.34.mlp.shared_expert.down_proj",
          "model.language_model.layers.34.mlp.shared_expert.gate_proj",
          "model.language_model.layers.34.mlp.shared_expert.up_proj",
          "model.language_model.layers.35.mlp.experts",
          "model.language_model.layers.35.mlp.shared_expert.down_proj",
          "model.language_model.layers.35.mlp.shared_expert.gate_proj",
          "model.language_model.layers.35.mlp.shared_expert.up_proj",
          "model.language_model.layers.36.mlp.experts",
          "model.language_model.layers.36.mlp.shared_expert.down_proj",
          "model.language_model.layers.36.mlp.shared_expert.gate_proj",
          "model.language_model.layers.36.mlp.shared_expert.up_proj",
          "model.language_model.layers.37.mlp.experts",
          "model.language_model.layers.37.mlp.shared_expert.down_proj",
          "model.language_model.layers.37.mlp.shared_expert.gate_proj",
          "model.language_model.layers.37.mlp.shared_expert.up_proj",
          "model.language_model.layers.38.mlp.experts",
          "model.language_model.layers.38.mlp.shared_expert.down_proj",
          "model.language_model.layers.38.mlp.shared_expert.gate_proj",
          "model.language_model.layers.38.mlp.shared_expert.up_proj",
          "model.language_model.layers.39.mlp.experts",
          "model.language_model.layers.39.mlp.shared_expert.down_proj",
          "model.language_model.layers.39.mlp.shared_expert.gate_proj",
          "model.language_model.layers.39.mlp.shared_expert.up_proj",
          "model.language_model.layers.4.mlp.experts",
          "model.language_model.layers.4.mlp.shared_expert.down_proj",
          "model.language_model.layers.4.mlp.shared_expert.gate_proj",
          "model.language_model.layers.4.mlp.shared_expert.up_proj",
          "model.language_model.layers.5.mlp.experts",
          "model.language_model.layers.5.mlp.shared_expert.down_proj",
          "model.language_model.layers.5.mlp.shared_expert.gate_proj",
          "model.language_model.layers.5.mlp.shared_expert.up_proj",
          "model.language_model.layers.6.mlp.experts",
          "model.language_model.layers.6.mlp.shared_expert.down_proj",
          "model.language_model.layers.6.mlp.shared_expert.gate_proj",
          "model.language_model.layers.6.mlp.shared_expert.up_proj",
          "model.language_model.layers.7.mlp.experts",
          "model.language_model.layers.7.mlp.shared_expert.down_proj",
          "model.language_model.layers.7.mlp.shared_expert.gate_proj",
          "model.language_model.layers.7.mlp.shared_expert.up_proj",
          "model.language_model.layers.8.mlp.experts",
          "model.language_model.layers.8.mlp.shared_expert.down_proj",
          "model.language_model.layers.8.mlp.shared_expert.gate_proj",
          "model.language_model.layers.8.mlp.shared_expert.up_proj",
          "model.language_model.layers.9.mlp.experts",
          "model.language_model.layers.9.mlp.shared_expert.down_proj",
          "model.language_model.layers.9.mlp.shared_expert.gate_proj",
          "model.language_model.layers.9.mlp.shared_expert.up_proj"
        ]
      }
    },
    "quantized_layers": {
      "model.language_model.layers.0.linear_attn.out_proj": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.0.linear_attn.in_proj_qkv": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.0.linear_attn.in_proj_z": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.0.mlp.experts": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.0.mlp.shared_expert.gate_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.0.mlp.shared_expert.up_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.0.mlp.shared_expert.down_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.1.linear_attn.out_proj": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.1.linear_attn.in_proj_qkv": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.1.linear_attn.in_proj_z": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.1.mlp.experts": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.1.mlp.shared_expert.gate_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.1.mlp.shared_expert.up_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.1.mlp.shared_expert.down_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.2.linear_attn.out_proj": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.2.linear_attn.in_proj_qkv": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.2.linear_attn.in_proj_z": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.2.mlp.experts": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.2.mlp.shared_expert.gate_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.2.mlp.shared_expert.up_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.2.mlp.shared_expert.down_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.3.self_attn.q_proj": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.3.self_attn.k_proj": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.3.self_attn.v_proj": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.3.self_attn.o_proj": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.3.mlp.experts": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.3.mlp.shared_expert.gate_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.3.mlp.shared_expert.up_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.3.mlp.shared_expert.down_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.4.linear_attn.out_proj": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.4.linear_attn.in_proj_qkv": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.4.linear_attn.in_proj_z": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.4.mlp.experts": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.4.mlp.shared_expert.gate_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.4.mlp.shared_expert.up_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.4.mlp.shared_expert.down_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.5.linear_attn.out_proj": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.5.linear_attn.in_proj_qkv": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.5.linear_attn.in_proj_z": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.5.mlp.experts": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.5.mlp.shared_expert.gate_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.5.mlp.shared_expert.up_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.5.mlp.shared_expert.down_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.6.linear_attn.out_proj": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.6.linear_attn.in_proj_qkv": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.6.linear_attn.in_proj_z": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.6.mlp.experts": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.6.mlp.shared_expert.gate_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.6.mlp.shared_expert.up_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.6.mlp.shared_expert.down_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.7.self_attn.q_proj": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.7.self_attn.k_proj": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.7.self_attn.v_proj": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.7.self_attn.o_proj": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.7.mlp.experts": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.7.mlp.shared_expert.gate_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.7.mlp.shared_expert.up_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.7.mlp.shared_expert.down_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.8.linear_attn.out_proj": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.8.linear_attn.in_proj_qkv": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.8.linear_attn.in_proj_z": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.8.mlp.experts": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.8.mlp.shared_expert.gate_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.8.mlp.shared_expert.up_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.8.mlp.shared_expert.down_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.9.linear_attn.out_proj": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.9.linear_attn.in_proj_qkv": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.9.linear_attn.in_proj_z": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.9.mlp.experts": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.9.mlp.shared_expert.gate_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.9.mlp.shared_expert.up_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.9.mlp.shared_expert.down_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.10.linear_attn.out_proj": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.10.linear_attn.in_proj_qkv": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.10.linear_attn.in_proj_z": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.10.mlp.experts": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.10.mlp.shared_expert.gate_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.10.mlp.shared_expert.up_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.10.mlp.shared_expert.down_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.11.self_attn.q_proj": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.11.self_attn.k_proj": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.11.self_attn.v_proj": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.11.self_attn.o_proj": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.11.mlp.experts": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.11.mlp.shared_expert.gate_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.11.mlp.shared_expert.up_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.11.mlp.shared_expert.down_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.12.linear_attn.out_proj": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.12.linear_attn.in_proj_qkv": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.12.linear_attn.in_proj_z": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.12.mlp.experts": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.12.mlp.shared_expert.gate_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.12.mlp.shared_expert.up_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.12.mlp.shared_expert.down_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.13.linear_attn.out_proj": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.13.linear_attn.in_proj_qkv": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.13.linear_attn.in_proj_z": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.13.mlp.experts": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.13.mlp.shared_expert.gate_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.13.mlp.shared_expert.up_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.13.mlp.shared_expert.down_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.14.linear_attn.out_proj": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.14.linear_attn.in_proj_qkv": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.14.linear_attn.in_proj_z": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.14.mlp.experts": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.14.mlp.shared_expert.gate_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.14.mlp.shared_expert.up_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.14.mlp.shared_expert.down_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.15.self_attn.q_proj": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.15.self_attn.k_proj": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.15.self_attn.v_proj": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.15.self_attn.o_proj": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.15.mlp.experts": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.15.mlp.shared_expert.gate_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.15.mlp.shared_expert.up_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.15.mlp.shared_expert.down_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.16.linear_attn.out_proj": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.16.linear_attn.in_proj_qkv": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.16.linear_attn.in_proj_z": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.16.mlp.experts": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.16.mlp.shared_expert.gate_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.16.mlp.shared_expert.up_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.16.mlp.shared_expert.down_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.17.linear_attn.out_proj": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.17.linear_attn.in_proj_qkv": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.17.linear_attn.in_proj_z": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.17.mlp.experts": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.17.mlp.shared_expert.gate_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.17.mlp.shared_expert.up_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.17.mlp.shared_expert.down_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.18.linear_attn.out_proj": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.18.linear_attn.in_proj_qkv": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.18.linear_attn.in_proj_z": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.18.mlp.experts": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.18.mlp.shared_expert.gate_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.18.mlp.shared_expert.up_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.18.mlp.shared_expert.down_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.19.self_attn.q_proj": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.19.self_attn.k_proj": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.19.self_attn.v_proj": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.19.self_attn.o_proj": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.19.mlp.experts": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.19.mlp.shared_expert.gate_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.19.mlp.shared_expert.up_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.19.mlp.shared_expert.down_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.20.linear_attn.out_proj": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.20.linear_attn.in_proj_qkv": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.20.linear_attn.in_proj_z": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.20.mlp.experts": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.20.mlp.shared_expert.gate_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.20.mlp.shared_expert.up_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.20.mlp.shared_expert.down_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.21.linear_attn.out_proj": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.21.linear_attn.in_proj_qkv": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.21.linear_attn.in_proj_z": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.21.mlp.experts": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.21.mlp.shared_expert.gate_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.21.mlp.shared_expert.up_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.21.mlp.shared_expert.down_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.22.linear_attn.out_proj": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.22.linear_attn.in_proj_qkv": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.22.linear_attn.in_proj_z": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.22.mlp.experts": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.22.mlp.shared_expert.gate_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.22.mlp.shared_expert.up_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.22.mlp.shared_expert.down_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.23.self_attn.q_proj": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.23.self_attn.k_proj": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.23.self_attn.v_proj": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.23.self_attn.o_proj": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.23.mlp.experts": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.23.mlp.shared_expert.gate_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.23.mlp.shared_expert.up_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.23.mlp.shared_expert.down_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.24.linear_attn.out_proj": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.24.linear_attn.in_proj_qkv": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.24.linear_attn.in_proj_z": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.24.mlp.experts": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.24.mlp.shared_expert.gate_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.24.mlp.shared_expert.up_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.24.mlp.shared_expert.down_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.25.linear_attn.out_proj": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.25.linear_attn.in_proj_qkv": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.25.linear_attn.in_proj_z": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.25.mlp.experts": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.25.mlp.shared_expert.gate_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.25.mlp.shared_expert.up_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.25.mlp.shared_expert.down_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.26.linear_attn.out_proj": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.26.linear_attn.in_proj_qkv": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.26.linear_attn.in_proj_z": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.26.mlp.experts": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.26.mlp.shared_expert.gate_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.26.mlp.shared_expert.up_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.26.mlp.shared_expert.down_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.27.self_attn.q_proj": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.27.self_attn.k_proj": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.27.self_attn.v_proj": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.27.self_attn.o_proj": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.27.mlp.experts": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.27.mlp.shared_expert.gate_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.27.mlp.shared_expert.up_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.27.mlp.shared_expert.down_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.28.linear_attn.out_proj": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.28.linear_attn.in_proj_qkv": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.28.linear_attn.in_proj_z": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.28.mlp.experts": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.28.mlp.shared_expert.gate_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.28.mlp.shared_expert.up_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.28.mlp.shared_expert.down_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.29.linear_attn.out_proj": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.29.linear_attn.in_proj_qkv": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.29.linear_attn.in_proj_z": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.29.mlp.experts": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.29.mlp.shared_expert.gate_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.29.mlp.shared_expert.up_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.29.mlp.shared_expert.down_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.30.linear_attn.out_proj": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.30.linear_attn.in_proj_qkv": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.30.linear_attn.in_proj_z": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.30.mlp.experts": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.30.mlp.shared_expert.gate_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.30.mlp.shared_expert.up_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.30.mlp.shared_expert.down_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.31.self_attn.q_proj": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.31.self_attn.k_proj": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.31.self_attn.v_proj": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.31.self_attn.o_proj": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.31.mlp.experts": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.31.mlp.shared_expert.gate_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.31.mlp.shared_expert.up_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.31.mlp.shared_expert.down_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.32.linear_attn.out_proj": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.32.linear_attn.in_proj_qkv": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.32.linear_attn.in_proj_z": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.32.mlp.experts": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.32.mlp.shared_expert.gate_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.32.mlp.shared_expert.up_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.32.mlp.shared_expert.down_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.33.linear_attn.out_proj": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.33.linear_attn.in_proj_qkv": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.33.linear_attn.in_proj_z": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.33.mlp.experts": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.33.mlp.shared_expert.gate_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.33.mlp.shared_expert.up_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.33.mlp.shared_expert.down_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.34.linear_attn.out_proj": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.34.linear_attn.in_proj_qkv": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.34.linear_attn.in_proj_z": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.34.mlp.experts": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.34.mlp.shared_expert.gate_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.34.mlp.shared_expert.up_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.34.mlp.shared_expert.down_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.35.self_attn.q_proj": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.35.self_attn.k_proj": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.35.self_attn.v_proj": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.35.self_attn.o_proj": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.35.mlp.experts": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.35.mlp.shared_expert.gate_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.35.mlp.shared_expert.up_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.35.mlp.shared_expert.down_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.36.linear_attn.out_proj": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.36.linear_attn.in_proj_qkv": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.36.linear_attn.in_proj_z": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.36.mlp.experts": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.36.mlp.shared_expert.gate_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.36.mlp.shared_expert.up_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.36.mlp.shared_expert.down_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.37.linear_attn.out_proj": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.37.linear_attn.in_proj_qkv": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.37.linear_attn.in_proj_z": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.37.mlp.experts": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.37.mlp.shared_expert.gate_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.37.mlp.shared_expert.up_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.37.mlp.shared_expert.down_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.38.linear_attn.out_proj": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.38.linear_attn.in_proj_qkv": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.38.linear_attn.in_proj_z": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.38.mlp.experts": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.38.mlp.shared_expert.gate_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.38.mlp.shared_expert.up_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.38.mlp.shared_expert.down_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.39.self_attn.q_proj": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.39.self_attn.k_proj": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.39.self_attn.v_proj": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.39.self_attn.o_proj": {
        "quant_algo": "FP8"
      },
      "model.language_model.layers.39.mlp.experts": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.39.mlp.shared_expert.gate_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.39.mlp.shared_expert.up_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "model.language_model.layers.39.mlp.shared_expert.down_proj": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      },
      "lm_head": {
        "quant_algo": "W4A16_NVFP4",
        "group_size": 16
      }
    },
    "ignore": [
      "mtp.layers.0*",
      "mtp*"
    ],
    "quant_algo": "MIXED_PRECISION",
    "producer": {
      "name": "modelopt",
      "version": "0.37.0"
    },
    "quant_method": "modelopt"
  }
}
