{ "architectures": [ "Glm5vForConditionalGeneration" ], "model_type": "glm5v", "text_config": { "architectures": [ "GlmMoeDsaForCausalLM" ], "attention_bias": false, "attention_dropout": 0.0, "dtype": "bfloat16", "eos_token_id": [ 154820, 154827, 154829 ], "ep_size": 1, "first_k_dense_replace": 3, "head_dim": 192, "hidden_act": "silu", "hidden_size": 6144, "index_head_dim": 128, "index_n_heads": 32, "index_share_for_mtp_iteration": true, "index_skip_topk_offset": 3, "index_topk": 2048, "index_topk_freq": 4, "index_topk_pattern": null, "indexer_rope_interleave": true, "indexer_types": [ "full", "full", "full", "shared", "shared", "shared", "full", "shared", "shared", "shared", "full", "shared", "shared", "shared", "full", "shared", "shared", "shared", "full", "shared", "shared", "shared", "full", "shared", "shared", "shared", "full", "shared", "shared", "shared", "full", "shared", "shared", "shared", "full", "shared", "shared", "shared", "full", "shared", "shared", "shared", "full", "shared", "shared", "shared", "full", "shared", "shared", "shared", "full", "shared", "shared", "shared", "full", "shared", "shared", "shared", "full", "shared", "shared", "shared", "full", "shared", "shared", "shared", "full", "shared", "shared", "shared", "full", "shared", "shared", "shared", "full", "shared", "shared", "shared" ], "initializer_range": 0.02, "intermediate_size": 12288, "kv_lora_rank": 512, "max_position_embeddings": 1048576, "mlp_layer_types": [ "dense", "dense", "dense", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse" ], "model_type": "glm_moe_dsa", "moe_intermediate_size": 2048, "moe_layer_freq": 1, "n_group": 1, "n_routed_experts": 256, "n_shared_experts": 1, "norm_topk_prob": true, "num_attention_heads": 64, "num_experts_per_tok": 8, "num_hidden_layers": 78, "num_key_value_heads": 64, "num_nextn_predict_layers": 1, "pad_token_id": 154820, "pretraining_tp": 1, "q_lora_rank": 2048, "qk_head_dim": 256, "qk_nope_head_dim": 192, "qk_rope_head_dim": 64, "rms_norm_eps": 1e-05, "rope_interleave": true, "rope_parameters": { "rope_theta": 8000000, "rope_type": "default" }, "routed_scaling_factor": 2.5, "scoring_func": "sigmoid", "tie_word_embeddings": false, "topk_group": 1, "topk_method": "noaux_tc", "transformers_version": "5.12.0", "use_cache": true, "v_head_dim": 256, "vocab_size": 154880, "quantization_config": { "quant_method": "nvfp4_aqlm_hybrid", "nvfp4": { "config_groups": { "group_0": { "input_activations": { "dynamic": false, "num_bits": 4, "type": "float", "group_size": 16 }, "weights": { "dynamic": false, "num_bits": 4, "type": "float", "group_size": 16 }, "targets": [ "Linear" ] } }, "ignore": [ "lm_head", "model.layers.0.mlp*", "model.layers.0.self_attn*", "model.layers.0.self_attn.indexer*", "model.layers.1.mlp*", "model.layers.1.self_attn*", "model.layers.1.self_attn.indexer*", "*shared_experts*", "model.layers.10.self_attn*", "model.layers.10.self_attn.indexer*", "model.layers.11.self_attn*", "model.layers.11.self_attn.indexer*", "model.layers.12.self_attn*", "model.layers.12.self_attn.indexer*", "model.layers.13.self_attn*", "model.layers.13.self_attn.indexer*", "model.layers.14.self_attn*", "model.layers.14.self_attn.indexer*", "model.layers.15.self_attn*", "model.layers.15.self_attn.indexer*", "model.layers.16.self_attn*", "model.layers.16.self_attn.indexer*", "model.layers.17.self_attn*", "model.layers.17.self_attn.indexer*", "model.layers.18.self_attn*", "model.layers.18.self_attn.indexer*", "model.layers.19.self_attn*", "model.layers.19.self_attn.indexer*", "model.layers.2.mlp*", "model.layers.2.self_attn*", "model.layers.2.self_attn.indexer*", "model.layers.20.self_attn*", "model.layers.20.self_attn.indexer*", "model.layers.21.self_attn*", "model.layers.21.self_attn.indexer*", "model.layers.22.self_attn*", "model.layers.22.self_attn.indexer*", "model.layers.23.self_attn*", "model.layers.23.self_attn.indexer*", "model.layers.24.self_attn*", "model.layers.24.self_attn.indexer*", "model.layers.25.self_attn*", "model.layers.25.self_attn.indexer*", "model.layers.26.self_attn*", "model.layers.26.self_attn.indexer*", "model.layers.27.self_attn*", "model.layers.27.self_attn.indexer*", "model.layers.28.self_attn*", "model.layers.28.self_attn.indexer*", "model.layers.29.self_attn*", "model.layers.29.self_attn.indexer*", "model.layers.3.self_attn*", "model.layers.3.self_attn.indexer*", "model.layers.30.self_attn*", "model.layers.30.self_attn.indexer*", "model.layers.31.self_attn*", "model.layers.31.self_attn.indexer*", "model.layers.32.self_attn*", "model.layers.32.self_attn.indexer*", "model.layers.33.self_attn*", "model.layers.33.self_attn.indexer*", "model.layers.34.self_attn*", "model.layers.34.self_attn.indexer*", "model.layers.35.self_attn*", "model.layers.35.self_attn.indexer*", "model.layers.36.self_attn*", "model.layers.36.self_attn.indexer*", "model.layers.37.self_attn*", "model.layers.37.self_attn.indexer*", "model.layers.38.self_attn*", "model.layers.38.self_attn.indexer*", "model.layers.39.self_attn*", "model.layers.39.self_attn.indexer*", "model.layers.4.self_attn*", "model.layers.4.self_attn.indexer*", "model.layers.40.self_attn*", "model.layers.40.self_attn.indexer*", "model.layers.41.self_attn*", "model.layers.41.self_attn.indexer*", "model.layers.42.self_attn*", "model.layers.42.self_attn.indexer*", "model.layers.43.self_attn*", "model.layers.43.self_attn.indexer*", "model.layers.44.self_attn*", "model.layers.44.self_attn.indexer*", "model.layers.45.self_attn*", "model.layers.45.self_attn.indexer*", "model.layers.46.self_attn*", "model.layers.46.self_attn.indexer*", "model.layers.47.self_attn*", "model.layers.47.self_attn.indexer*", "model.layers.48.self_attn*", "model.layers.48.self_attn.indexer*", "model.layers.49.self_attn*", "model.layers.49.self_attn.indexer*", "model.layers.5.self_attn*", "model.layers.5.self_attn.indexer*", "model.layers.50.self_attn*", "model.layers.50.self_attn.indexer*", "model.layers.51.self_attn*", "model.layers.51.self_attn.indexer*", "model.layers.52.self_attn*", "model.layers.52.self_attn.indexer*", "model.layers.53.self_attn*", "model.layers.53.self_attn.indexer*", "model.layers.54.self_attn*", "model.layers.54.self_attn.indexer*", "model.layers.55.self_attn*", "model.layers.55.self_attn.indexer*", "model.layers.56.self_attn*", "model.layers.56.self_attn.indexer*", "model.layers.57.self_attn*", "model.layers.57.self_attn.indexer*", "model.layers.58.self_attn*", "model.layers.58.self_attn.indexer*", "model.layers.59.self_attn*", "model.layers.59.self_attn.indexer*", "model.layers.6.self_attn*", "model.layers.6.self_attn.indexer*", "model.layers.60.self_attn*", "model.layers.60.self_attn.indexer*", "model.layers.61.self_attn*", "model.layers.61.self_attn.indexer*", "model.layers.62.self_attn*", "model.layers.62.self_attn.indexer*", "model.layers.63.self_attn*", "model.layers.63.self_attn.indexer*", "model.layers.64.self_attn*", "model.layers.64.self_attn.indexer*", "model.layers.65.self_attn*", "model.layers.65.self_attn.indexer*", "model.layers.66.self_attn*", "model.layers.66.self_attn.indexer*", "model.layers.67.self_attn*", "model.layers.67.self_attn.indexer*", "model.layers.68.self_attn*", "model.layers.68.self_attn.indexer*", "model.layers.69.self_attn*", "model.layers.69.self_attn.indexer*", "model.layers.7.self_attn*", "model.layers.7.self_attn.indexer*", "model.layers.70.self_attn*", "model.layers.70.self_attn.indexer*", "model.layers.71.self_attn*", "model.layers.71.self_attn.indexer*", "model.layers.72.self_attn*", "model.layers.72.self_attn.indexer*", "model.layers.73.self_attn*", "model.layers.73.self_attn.indexer*", "model.layers.74.self_attn*", "model.layers.74.self_attn.indexer*", "model.layers.75.self_attn*", "model.layers.75.self_attn.indexer*", "model.layers.76.self_attn*", "model.layers.76.self_attn.indexer*", "model.layers.77.self_attn*", "model.layers.77.self_attn.indexer*", "model.layers.8.self_attn*", "model.layers.8.self_attn.indexer*", "model.layers.9.self_attn*", "model.layers.9.self_attn.indexer*", "model.layers.78.eh_proj*", "model.layers.78.self_attn*", "model.layers.78.self_attn.indexer*", "model.layers.78.mlp.gate*", "model.layers.78.mlp.shared_experts*" ], "quant_algo": "NVFP4", "producer": { "name": "modelopt", "version": "0.39.0.dev290+gf9d9a71de.d20260214" }, "quant_method": "modelopt" }, "aqlm": { "entries": 65536, "group_size": 8 }, "aqlm_layer_books": { "3": { "n_nvfp4": 29, "n_base": 0, "n_cold": 227 }, "4": { "n_nvfp4": 48, "n_base": 0, "n_cold": 208 }, "5": { "n_nvfp4": 33, "n_base": 0, "n_cold": 223 }, "6": { "n_nvfp4": 38, "n_base": 0, "n_cold": 218 }, "7": { "n_nvfp4": 39, "n_base": 0, "n_cold": 217 }, "8": { "n_nvfp4": 60, "n_base": 0, "n_cold": 196 }, "9": { "n_nvfp4": 66, "n_base": 0, "n_cold": 190 }, "10": { "n_nvfp4": 48, "n_base": 0, "n_cold": 208 }, "11": { "n_nvfp4": 50, "n_base": 0, "n_cold": 206 }, "12": { "n_nvfp4": 51, "n_base": 0, "n_cold": 205 }, "13": { "n_nvfp4": 58, "n_base": 0, "n_cold": 198 }, "14": { "n_nvfp4": 60, "n_base": 0, "n_cold": 196 }, "15": { "n_nvfp4": 51, "n_base": 0, "n_cold": 205 }, "16": { "n_nvfp4": 51, "n_base": 0, "n_cold": 205 }, "17": { "n_nvfp4": 39, "n_base": 0, "n_cold": 217 }, "18": { "n_nvfp4": 37, "n_base": 0, "n_cold": 219 }, "19": { "n_nvfp4": 43, "n_base": 0, "n_cold": 213 }, "20": { "n_nvfp4": 33, "n_base": 0, "n_cold": 223 }, "21": { "n_nvfp4": 32, "n_base": 0, "n_cold": 224 }, "22": { "n_nvfp4": 35, "n_base": 0, "n_cold": 221 }, "23": { "n_nvfp4": 59, "n_base": 0, "n_cold": 197 }, "24": { "n_nvfp4": 61, "n_base": 0, "n_cold": 195 }, "25": { "n_nvfp4": 48, "n_base": 0, "n_cold": 208 }, "26": { "n_nvfp4": 55, "n_base": 0, "n_cold": 201 }, "27": { "n_nvfp4": 49, "n_base": 0, "n_cold": 207 }, "28": { "n_nvfp4": 64, "n_base": 0, "n_cold": 192 }, "29": { "n_nvfp4": 65, "n_base": 0, "n_cold": 191 }, "30": { "n_nvfp4": 84, "n_base": 0, "n_cold": 172 }, "31": { "n_nvfp4": 77, "n_base": 0, "n_cold": 179 }, "32": { "n_nvfp4": 64, "n_base": 0, "n_cold": 192 }, "33": { "n_nvfp4": 78, "n_base": 0, "n_cold": 178 }, "34": { "n_nvfp4": 89, "n_base": 0, "n_cold": 167 }, "35": { "n_nvfp4": 85, "n_base": 0, "n_cold": 171 }, "36": { "n_nvfp4": 79, "n_base": 0, "n_cold": 177 }, "37": { "n_nvfp4": 76, "n_base": 0, "n_cold": 180 }, "38": { "n_nvfp4": 79, "n_base": 0, "n_cold": 177 }, "39": { "n_nvfp4": 80, "n_base": 0, "n_cold": 176 }, "40": { "n_nvfp4": 86, "n_base": 0, "n_cold": 170 }, "41": { "n_nvfp4": 90, "n_base": 0, "n_cold": 166 }, "42": { "n_nvfp4": 84, "n_base": 0, "n_cold": 172 }, "43": { "n_nvfp4": 73, "n_base": 0, "n_cold": 183 }, "44": { "n_nvfp4": 90, "n_base": 0, "n_cold": 166 }, "45": { "n_nvfp4": 73, "n_base": 0, "n_cold": 183 }, "46": { "n_nvfp4": 76, "n_base": 0, "n_cold": 180 }, "47": { "n_nvfp4": 83, "n_base": 0, "n_cold": 173 }, "48": { "n_nvfp4": 72, "n_base": 0, "n_cold": 184 }, "49": { "n_nvfp4": 75, "n_base": 0, "n_cold": 181 }, "50": { "n_nvfp4": 61, "n_base": 0, "n_cold": 195 }, "51": { "n_nvfp4": 61, "n_base": 0, "n_cold": 195 }, "52": { "n_nvfp4": 68, "n_base": 0, "n_cold": 188 }, "53": { "n_nvfp4": 65, "n_base": 0, "n_cold": 191 }, "54": { "n_nvfp4": 61, "n_base": 0, "n_cold": 195 }, "55": { "n_nvfp4": 62, "n_base": 0, "n_cold": 194 }, "56": { "n_nvfp4": 73, "n_base": 0, "n_cold": 183 }, "57": { "n_nvfp4": 80, "n_base": 0, "n_cold": 176 }, "58": { "n_nvfp4": 65, "n_base": 0, "n_cold": 191 }, "59": { "n_nvfp4": 69, "n_base": 0, "n_cold": 187 }, "60": { "n_nvfp4": 63, "n_base": 0, "n_cold": 193 }, "61": { "n_nvfp4": 92, "n_base": 0, "n_cold": 164 }, "62": { "n_nvfp4": 96, "n_base": 0, "n_cold": 160 }, "63": { "n_nvfp4": 82, "n_base": 0, "n_cold": 174 }, "64": { "n_nvfp4": 93, "n_base": 0, "n_cold": 163 }, "65": { "n_nvfp4": 95, "n_base": 0, "n_cold": 161 }, "66": { "n_nvfp4": 104, "n_base": 0, "n_cold": 152 }, "67": { "n_nvfp4": 93, "n_base": 0, "n_cold": 163 }, "68": { "n_nvfp4": 123, "n_base": 0, "n_cold": 133 }, "69": { "n_nvfp4": 127, "n_base": 0, "n_cold": 129 }, "70": { "n_nvfp4": 144, "n_base": 0, "n_cold": 112 }, "71": { "n_nvfp4": 137, "n_base": 0, "n_cold": 119 }, "72": { "n_nvfp4": 135, "n_base": 0, "n_cold": 121 }, "73": { "n_nvfp4": 138, "n_base": 0, "n_cold": 118 }, "74": { "n_nvfp4": 142, "n_base": 0, "n_cold": 114 }, "75": { "n_nvfp4": 140, "n_base": 0, "n_cold": 116 }, "76": { "n_nvfp4": 144, "n_base": 0, "n_cold": 112 }, "77": { "n_nvfp4": 134, "n_base": 0, "n_cold": 122 } } } }, "vision_config": { "patch_size": 14, "init_pos_emb_height": 64, "init_pos_emb_width": 64, "init_pos_emb_time": 4, "pos_emb_type": "divided_fixed", "num_attention_heads": 16, "num_hidden_layers": 27, "hidden_size": 1152, "intermediate_size": 4304, "vt_num_attention_heads": 16, "vt_num_hidden_layers": 27, "vt_hidden_size": 1152, "vt_intermediate_size": 4304, "merge_kernel_size": [ 2, 2 ], "video_attn_type": "spatial_temporal", "merge_type": "sd2_tpool", "mm_projector_type": "patchmerger", "mm_hidden_size": 1152, "projector_hidden_act": "gelu", "projector_ln_eps": 1e-05, "text_hidden_size": 6144 }, "ignore_index": -100, "media_placeholder_token_id": 154854, "pad_token_id": 154820, "use_unified_vision_chunk": true, "video_placeholder": "<|glm5v_video_placeholder|>", "encoder_only": false, "language_only": false, "tie_word_embeddings": false, "dtype": "bfloat16", "quantization_config": { "quant_method": "nvfp4_aqlm_hybrid", "nvfp4": { "config_groups": { "group_0": { "input_activations": { "dynamic": false, "num_bits": 4, "type": "float", "group_size": 16 }, "weights": { "dynamic": false, "num_bits": 4, "type": "float", "group_size": 16 }, "targets": [ "Linear" ] } }, "ignore": [ "lm_head", "model.layers.0.mlp*", "model.layers.0.self_attn*", "model.layers.0.self_attn.indexer*", "model.layers.1.mlp*", "model.layers.1.self_attn*", "model.layers.1.self_attn.indexer*", "*shared_experts*", "model.layers.10.self_attn*", "model.layers.10.self_attn.indexer*", "model.layers.11.self_attn*", "model.layers.11.self_attn.indexer*", "model.layers.12.self_attn*", "model.layers.12.self_attn.indexer*", "model.layers.13.self_attn*", "model.layers.13.self_attn.indexer*", "model.layers.14.self_attn*", "model.layers.14.self_attn.indexer*", "model.layers.15.self_attn*", "model.layers.15.self_attn.indexer*", "model.layers.16.self_attn*", "model.layers.16.self_attn.indexer*", "model.layers.17.self_attn*", "model.layers.17.self_attn.indexer*", "model.layers.18.self_attn*", "model.layers.18.self_attn.indexer*", "model.layers.19.self_attn*", "model.layers.19.self_attn.indexer*", "model.layers.2.mlp*", "model.layers.2.self_attn*", "model.layers.2.self_attn.indexer*", "model.layers.20.self_attn*", "model.layers.20.self_attn.indexer*", "model.layers.21.self_attn*", "model.layers.21.self_attn.indexer*", "model.layers.22.self_attn*", "model.layers.22.self_attn.indexer*", "model.layers.23.self_attn*", "model.layers.23.self_attn.indexer*", "model.layers.24.self_attn*", "model.layers.24.self_attn.indexer*", "model.layers.25.self_attn*", "model.layers.25.self_attn.indexer*", "model.layers.26.self_attn*", "model.layers.26.self_attn.indexer*", "model.layers.27.self_attn*", "model.layers.27.self_attn.indexer*", "model.layers.28.self_attn*", "model.layers.28.self_attn.indexer*", "model.layers.29.self_attn*", "model.layers.29.self_attn.indexer*", "model.layers.3.self_attn*", "model.layers.3.self_attn.indexer*", "model.layers.30.self_attn*", "model.layers.30.self_attn.indexer*", "model.layers.31.self_attn*", "model.layers.31.self_attn.indexer*", "model.layers.32.self_attn*", "model.layers.32.self_attn.indexer*", "model.layers.33.self_attn*", "model.layers.33.self_attn.indexer*", "model.layers.34.self_attn*", "model.layers.34.self_attn.indexer*", "model.layers.35.self_attn*", "model.layers.35.self_attn.indexer*", "model.layers.36.self_attn*", "model.layers.36.self_attn.indexer*", "model.layers.37.self_attn*", "model.layers.37.self_attn.indexer*", "model.layers.38.self_attn*", "model.layers.38.self_attn.indexer*", "model.layers.39.self_attn*", "model.layers.39.self_attn.indexer*", "model.layers.4.self_attn*", "model.layers.4.self_attn.indexer*", "model.layers.40.self_attn*", "model.layers.40.self_attn.indexer*", "model.layers.41.self_attn*", "model.layers.41.self_attn.indexer*", "model.layers.42.self_attn*", "model.layers.42.self_attn.indexer*", "model.layers.43.self_attn*", "model.layers.43.self_attn.indexer*", "model.layers.44.self_attn*", "model.layers.44.self_attn.indexer*", "model.layers.45.self_attn*", "model.layers.45.self_attn.indexer*", "model.layers.46.self_attn*", "model.layers.46.self_attn.indexer*", "model.layers.47.self_attn*", "model.layers.47.self_attn.indexer*", "model.layers.48.self_attn*", "model.layers.48.self_attn.indexer*", "model.layers.49.self_attn*", "model.layers.49.self_attn.indexer*", "model.layers.5.self_attn*", "model.layers.5.self_attn.indexer*", "model.layers.50.self_attn*", "model.layers.50.self_attn.indexer*", "model.layers.51.self_attn*", "model.layers.51.self_attn.indexer*", "model.layers.52.self_attn*", "model.layers.52.self_attn.indexer*", "model.layers.53.self_attn*", "model.layers.53.self_attn.indexer*", "model.layers.54.self_attn*", "model.layers.54.self_attn.indexer*", "model.layers.55.self_attn*", "model.layers.55.self_attn.indexer*", "model.layers.56.self_attn*", "model.layers.56.self_attn.indexer*", "model.layers.57.self_attn*", "model.layers.57.self_attn.indexer*", "model.layers.58.self_attn*", "model.layers.58.self_attn.indexer*", "model.layers.59.self_attn*", "model.layers.59.self_attn.indexer*", "model.layers.6.self_attn*", "model.layers.6.self_attn.indexer*", "model.layers.60.self_attn*", "model.layers.60.self_attn.indexer*", "model.layers.61.self_attn*", "model.layers.61.self_attn.indexer*", "model.layers.62.self_attn*", "model.layers.62.self_attn.indexer*", "model.layers.63.self_attn*", "model.layers.63.self_attn.indexer*", "model.layers.64.self_attn*", "model.layers.64.self_attn.indexer*", "model.layers.65.self_attn*", "model.layers.65.self_attn.indexer*", "model.layers.66.self_attn*", "model.layers.66.self_attn.indexer*", "model.layers.67.self_attn*", "model.layers.67.self_attn.indexer*", "model.layers.68.self_attn*", "model.layers.68.self_attn.indexer*", "model.layers.69.self_attn*", "model.layers.69.self_attn.indexer*", "model.layers.7.self_attn*", "model.layers.7.self_attn.indexer*", "model.layers.70.self_attn*", "model.layers.70.self_attn.indexer*", "model.layers.71.self_attn*", "model.layers.71.self_attn.indexer*", "model.layers.72.self_attn*", "model.layers.72.self_attn.indexer*", "model.layers.73.self_attn*", "model.layers.73.self_attn.indexer*", "model.layers.74.self_attn*", "model.layers.74.self_attn.indexer*", "model.layers.75.self_attn*", "model.layers.75.self_attn.indexer*", "model.layers.76.self_attn*", "model.layers.76.self_attn.indexer*", "model.layers.77.self_attn*", "model.layers.77.self_attn.indexer*", "model.layers.8.self_attn*", "model.layers.8.self_attn.indexer*", "model.layers.9.self_attn*", "model.layers.9.self_attn.indexer*", "model.layers.78.eh_proj*", "model.layers.78.self_attn*", "model.layers.78.self_attn.indexer*", "model.layers.78.mlp.gate*", "model.layers.78.mlp.shared_experts*" ], "quant_algo": "NVFP4", "producer": { "name": "modelopt", "version": "0.39.0.dev290+gf9d9a71de.d20260214" }, "quant_method": "modelopt" }, "aqlm": { "entries": 65536, "group_size": 8 }, "aqlm_layer_books": { "3": { "n_nvfp4": 29, "n_base": 0, "n_cold": 227 }, "4": { "n_nvfp4": 48, "n_base": 0, "n_cold": 208 }, "5": { "n_nvfp4": 33, "n_base": 0, "n_cold": 223 }, "6": { "n_nvfp4": 38, "n_base": 0, "n_cold": 218 }, "7": { "n_nvfp4": 39, "n_base": 0, "n_cold": 217 }, "8": { "n_nvfp4": 60, "n_base": 0, "n_cold": 196 }, "9": { "n_nvfp4": 66, "n_base": 0, "n_cold": 190 }, "10": { "n_nvfp4": 48, "n_base": 0, "n_cold": 208 }, "11": { "n_nvfp4": 50, "n_base": 0, "n_cold": 206 }, "12": { "n_nvfp4": 51, "n_base": 0, "n_cold": 205 }, "13": { "n_nvfp4": 58, "n_base": 0, "n_cold": 198 }, "14": { "n_nvfp4": 60, "n_base": 0, "n_cold": 196 }, "15": { "n_nvfp4": 51, "n_base": 0, "n_cold": 205 }, "16": { "n_nvfp4": 51, "n_base": 0, "n_cold": 205 }, "17": { "n_nvfp4": 39, "n_base": 0, "n_cold": 217 }, "18": { "n_nvfp4": 37, "n_base": 0, "n_cold": 219 }, "19": { "n_nvfp4": 43, "n_base": 0, "n_cold": 213 }, "20": { "n_nvfp4": 33, "n_base": 0, "n_cold": 223 }, "21": { "n_nvfp4": 32, "n_base": 0, "n_cold": 224 }, "22": { "n_nvfp4": 35, "n_base": 0, "n_cold": 221 }, "23": { "n_nvfp4": 59, "n_base": 0, "n_cold": 197 }, "24": { "n_nvfp4": 61, "n_base": 0, "n_cold": 195 }, "25": { "n_nvfp4": 48, "n_base": 0, "n_cold": 208 }, "26": { "n_nvfp4": 55, "n_base": 0, "n_cold": 201 }, "27": { "n_nvfp4": 49, "n_base": 0, "n_cold": 207 }, "28": { "n_nvfp4": 64, "n_base": 0, "n_cold": 192 }, "29": { "n_nvfp4": 65, "n_base": 0, "n_cold": 191 }, "30": { "n_nvfp4": 84, "n_base": 0, "n_cold": 172 }, "31": { "n_nvfp4": 77, "n_base": 0, "n_cold": 179 }, "32": { "n_nvfp4": 64, "n_base": 0, "n_cold": 192 }, "33": { "n_nvfp4": 78, "n_base": 0, "n_cold": 178 }, "34": { "n_nvfp4": 89, "n_base": 0, "n_cold": 167 }, "35": { "n_nvfp4": 85, "n_base": 0, "n_cold": 171 }, "36": { "n_nvfp4": 79, "n_base": 0, "n_cold": 177 }, "37": { "n_nvfp4": 76, "n_base": 0, "n_cold": 180 }, "38": { "n_nvfp4": 79, "n_base": 0, "n_cold": 177 }, "39": { "n_nvfp4": 80, "n_base": 0, "n_cold": 176 }, "40": { "n_nvfp4": 86, "n_base": 0, "n_cold": 170 }, "41": { "n_nvfp4": 90, "n_base": 0, "n_cold": 166 }, "42": { "n_nvfp4": 84, "n_base": 0, "n_cold": 172 }, "43": { "n_nvfp4": 73, "n_base": 0, "n_cold": 183 }, "44": { "n_nvfp4": 90, "n_base": 0, "n_cold": 166 }, "45": { "n_nvfp4": 73, "n_base": 0, "n_cold": 183 }, "46": { "n_nvfp4": 76, "n_base": 0, "n_cold": 180 }, "47": { "n_nvfp4": 83, "n_base": 0, "n_cold": 173 }, "48": { "n_nvfp4": 72, "n_base": 0, "n_cold": 184 }, "49": { "n_nvfp4": 75, "n_base": 0, "n_cold": 181 }, "50": { "n_nvfp4": 61, "n_base": 0, "n_cold": 195 }, "51": { "n_nvfp4": 61, "n_base": 0, "n_cold": 195 }, "52": { "n_nvfp4": 68, "n_base": 0, "n_cold": 188 }, "53": { "n_nvfp4": 65, "n_base": 0, "n_cold": 191 }, "54": { "n_nvfp4": 61, "n_base": 0, "n_cold": 195 }, "55": { "n_nvfp4": 62, "n_base": 0, "n_cold": 194 }, "56": { "n_nvfp4": 73, "n_base": 0, "n_cold": 183 }, "57": { "n_nvfp4": 80, "n_base": 0, "n_cold": 176 }, "58": { "n_nvfp4": 65, "n_base": 0, "n_cold": 191 }, "59": { "n_nvfp4": 69, "n_base": 0, "n_cold": 187 }, "60": { "n_nvfp4": 63, "n_base": 0, "n_cold": 193 }, "61": { "n_nvfp4": 92, "n_base": 0, "n_cold": 164 }, "62": { "n_nvfp4": 96, "n_base": 0, "n_cold": 160 }, "63": { "n_nvfp4": 82, "n_base": 0, "n_cold": 174 }, "64": { "n_nvfp4": 93, "n_base": 0, "n_cold": 163 }, "65": { "n_nvfp4": 95, "n_base": 0, "n_cold": 161 }, "66": { "n_nvfp4": 104, "n_base": 0, "n_cold": 152 }, "67": { "n_nvfp4": 93, "n_base": 0, "n_cold": 163 }, "68": { "n_nvfp4": 123, "n_base": 0, "n_cold": 133 }, "69": { "n_nvfp4": 127, "n_base": 0, "n_cold": 129 }, "70": { "n_nvfp4": 144, "n_base": 0, "n_cold": 112 }, "71": { "n_nvfp4": 137, "n_base": 0, "n_cold": 119 }, "72": { "n_nvfp4": 135, "n_base": 0, "n_cold": 121 }, "73": { "n_nvfp4": 138, "n_base": 0, "n_cold": 118 }, "74": { "n_nvfp4": 142, "n_base": 0, "n_cold": 114 }, "75": { "n_nvfp4": 140, "n_base": 0, "n_cold": 116 }, "76": { "n_nvfp4": 144, "n_base": 0, "n_cold": 112 }, "77": { "n_nvfp4": 134, "n_base": 0, "n_cold": 122 } } }, "eos_token_id": [ 154820, 154827, 154829 ] }