| { |
| "architectures": [ |
| "KimiK3ForConditionalGeneration" |
| ], |
| "auto_map": { |
| "AutoConfig": "configuration_kimi_k3.KimiK3Config", |
| "AutoModel": "modeling_kimi_k3.KimiK3ForConditionalGeneration", |
| "AutoModelForCausalLM": "modeling_kimi_k3.KimiK3ForConditionalGeneration" |
| }, |
| "bos_token_id": 163584, |
| "dtype": "bfloat16", |
| "eos_token_id": 163586, |
| "ignore_index": -100, |
| "image_placeholder": "<|kimi_image_placeholder|>", |
| "media_placeholder_token_id": 163605, |
| "model_type": "kimi_k3", |
| "pad_token_id": 163839, |
| "text_config": { |
| "_name_or_path": "", |
| "activation_situ_beta": 4.0, |
| "activation_situ_linear_beta": 25.0, |
| "add_cross_attention": false, |
| "architectures": [ |
| "KimiLinearForCausalLM" |
| ], |
| "attn_res_block_size": 12, |
| "auto_map": { |
| "AutoConfig": "configuration_kimi_k3.KimiLinearConfig", |
| "AutoModel": "modeling_kimi_linear.KimiLinearModel", |
| "AutoModelForCausalLM": "modeling_kimi_linear.KimiLinearForCausalLM" |
| }, |
| "bad_words_ids": null, |
| "begin_suppress_tokens": null, |
| "bos_token_id": 163584, |
| "chunk_size_feed_forward": 0, |
| "cross_attention_hidden_size": null, |
| "decoder_start_token_id": null, |
| "diversity_penalty": 0.0, |
| "do_sample": false, |
| "dtype": "bfloat16", |
| "early_stopping": false, |
| "encoder_no_repeat_ngram_size": 0, |
| "eos_token_id": 163586, |
| "exponential_decay_length_penalty": null, |
| "finetuning_task": null, |
| "first_k_dense_replace": 1, |
| "forced_bos_token_id": null, |
| "forced_eos_token_id": null, |
| "hidden_act": "situ", |
| "hidden_size": 7168, |
| "id2label": { |
| "0": "LABEL_0", |
| "1": "LABEL_1" |
| }, |
| "initializer_range": 0.02, |
| "intermediate_size": 33792, |
| "is_decoder": false, |
| "is_encoder_decoder": false, |
| "kv_lora_rank": 512, |
| "label2id": { |
| "LABEL_0": 0, |
| "LABEL_1": 1 |
| }, |
| "latent_moe_use_norm": true, |
| "length_penalty": 1.0, |
| "linear_attn_config": { |
| "full_attn_layers": [ |
| 4, |
| 8, |
| 12, |
| 16, |
| 20, |
| 24, |
| 28, |
| 32, |
| 36, |
| 40, |
| 44, |
| 48, |
| 52, |
| 56, |
| 60, |
| 64, |
| 68, |
| 72, |
| 76, |
| 80, |
| 84, |
| 88, |
| 92, |
| 93 |
| ], |
| "gate_lower_bound": -5.0, |
| "head_dim": 128, |
| "kda_layers": [ |
| 1, |
| 2, |
| 3, |
| 5, |
| 6, |
| 7, |
| 9, |
| 10, |
| 11, |
| 13, |
| 14, |
| 15, |
| 17, |
| 18, |
| 19, |
| 21, |
| 22, |
| 23, |
| 25, |
| 26, |
| 27, |
| 29, |
| 30, |
| 31, |
| 33, |
| 34, |
| 35, |
| 37, |
| 38, |
| 39, |
| 41, |
| 42, |
| 43, |
| 45, |
| 46, |
| 47, |
| 49, |
| 50, |
| 51, |
| 53, |
| 54, |
| 55, |
| 57, |
| 58, |
| 59, |
| 61, |
| 62, |
| 63, |
| 65, |
| 66, |
| 67, |
| 69, |
| 70, |
| 71, |
| 73, |
| 74, |
| 75, |
| 77, |
| 78, |
| 79, |
| 81, |
| 82, |
| 83, |
| 85, |
| 86, |
| 87, |
| 89, |
| 90, |
| 91 |
| ], |
| "num_heads": 96, |
| "short_conv_kernel_size": 4, |
| "use_full_rank_gate": true |
| }, |
| "max_length": 20, |
| "max_position_embeddings": 1048576, |
| "min_length": 0, |
| "mla_use_nope": true, |
| "mla_use_output_gate": true, |
| "model_type": "kimi_linear", |
| "moe_intermediate_size": 3072, |
| "moe_layer_freq": 1, |
| "moe_renormalize": true, |
| "moe_router_activation_func": "sigmoid", |
| "no_repeat_ngram_size": 0, |
| "num_attention_heads": 96, |
| "num_beam_groups": 1, |
| "num_beams": 1, |
| "num_expert_group": 1, |
| "num_experts": 896, |
| "num_experts_per_token": 16, |
| "num_hidden_layers": 93, |
| "num_key_value_heads": 96, |
| "num_nextn_predict_layers": 0, |
| "num_return_sequences": 1, |
| "num_shared_experts": 2, |
| "output_attentions": false, |
| "output_hidden_states": false, |
| "output_scores": false, |
| "pad_token_id": 163839, |
| "prefix": null, |
| "problem_type": null, |
| "pruned_heads": {}, |
| "q_lora_rank": 1536, |
| "qk_nope_head_dim": 128, |
| "qk_rope_head_dim": 64, |
| "quantization_config": { |
| "config_groups": { |
| "group_0": { |
| "format": "nvfp4-pack-quantized", |
| "input_activations": null, |
| "output_activations": null, |
| "targets": [ |
| "Linear" |
| ], |
| "weights": { |
| "actorder": null, |
| "block_structure": null, |
| "dynamic": false, |
| "group_size": 16, |
| "num_bits": 4, |
| "observer": "minmax", |
| "observer_kwargs": {}, |
| "scale_dtype": "torch.float8_e4m3fn", |
| "strategy": "tensor_group", |
| "symmetric": true, |
| "type": "float", |
| "zp_dtype": null |
| } |
| } |
| }, |
| "format": "nvfp4-pack-quantized", |
| "global_compression_ratio": null, |
| "ignore": [ |
| "*self_attn*", |
| "*shared_experts*", |
| "*mlp.gate_proj*", |
| "*mlp.up_proj*", |
| "*mlp.gate_up_proj*", |
| "*mlp.down_proj*", |
| "*lm_head*", |
| "*vision_tower*", |
| "*mm_projector*", |
| "*language_model.lm_head*", |
| "*language_model.model.embed_tokens*", |
| "*input_layernorm*", |
| "*mlp_res_norm*", |
| "*mlp_res_proj*", |
| "*post_attention_layernorm*", |
| "*self_attention_res_norm*", |
| "*self_attention_res_proj*", |
| "*self_attn.b_proj*", |
| "*self_attn.f_a_proj*", |
| "*self_attn.f_b_proj*", |
| "*self_attn.g_proj*", |
| "*self_attn.k_conv1d*", |
| "*self_attn.k_proj*", |
| "*self_attn.o_norm*", |
| "*self_attn.o_proj*", |
| "*self_attn.q_conv1d*", |
| "*self_attn.q_proj*", |
| "*self_attn.v_conv1d*", |
| "*self_attn.v_proj*", |
| "*block_sparse_moe.gate*", |
| "*block_sparse_moe.routed_expert_down_proj*", |
| "*block_sparse_moe.routed_expert_norm*", |
| "*block_sparse_moe.routed_expert_up_proj*", |
| "*block_sparse_moe.shared_experts.down_proj*", |
| "*block_sparse_moe.shared_experts.gate_proj*", |
| "*block_sparse_moe.shared_experts.up_proj*", |
| "*self_attn.kv_a_layernorm*", |
| "*self_attn.kv_a_proj_with_mqa*", |
| "*self_attn.kv_b_proj*", |
| "*self_attn.q_a_layernorm*", |
| "*self_attn.q_a_proj*", |
| "*self_attn.q_b_proj*", |
| "*language_model.model.norm*", |
| "*language_model.model.output_attn_res_norm*", |
| "*language_model.model.output_attn_res_proj*", |
| "*mm_projector.post_norm*", |
| "*mlp.fc0*", |
| "*mlp.fc1*", |
| "*norm0*", |
| "*norm1*", |
| "*wo*", |
| "*wqkv*", |
| "*vision_tower.encoder.final_layernorm*", |
| "*vision_tower.patch_embed.pos_emb*", |
| "*vision_tower.patch_embed.proj*" |
| ], |
| "kv_cache_scheme": null, |
| "quant_method": "modelopt", |
| "quantization_status": "compressed", |
| "group_size": 16, |
| "quant_algo": "NVFP4", |
| "kv_cache_quant_algo": null, |
| "exclude_modules": [ |
| "*self_attn*", |
| "*shared_experts*", |
| "*mlp.gate_proj*", |
| "*mlp.up_proj*", |
| "*mlp.gate_up_proj*", |
| "*mlp.down_proj*", |
| "*lm_head*", |
| "*vision_tower*", |
| "*mm_projector*", |
| "*language_model.lm_head*", |
| "*language_model.model.embed_tokens*", |
| "*input_layernorm*", |
| "*mlp_res_norm*", |
| "*mlp_res_proj*", |
| "*post_attention_layernorm*", |
| "*self_attention_res_norm*", |
| "*self_attention_res_proj*", |
| "*self_attn.b_proj*", |
| "*self_attn.f_a_proj*", |
| "*self_attn.f_b_proj*", |
| "*self_attn.g_proj*", |
| "*self_attn.k_conv1d*", |
| "*self_attn.k_proj*", |
| "*self_attn.o_norm*", |
| "*self_attn.o_proj*", |
| "*self_attn.q_conv1d*", |
| "*self_attn.q_proj*", |
| "*self_attn.v_conv1d*", |
| "*self_attn.v_proj*", |
| "*block_sparse_moe.gate*", |
| "*block_sparse_moe.routed_expert_down_proj*", |
| "*block_sparse_moe.routed_expert_norm*", |
| "*block_sparse_moe.routed_expert_up_proj*", |
| "*block_sparse_moe.shared_experts.down_proj*", |
| "*block_sparse_moe.shared_experts.gate_proj*", |
| "*block_sparse_moe.shared_experts.up_proj*", |
| "*self_attn.kv_a_layernorm*", |
| "*self_attn.kv_a_proj_with_mqa*", |
| "*self_attn.kv_b_proj*", |
| "*self_attn.q_a_layernorm*", |
| "*self_attn.q_a_proj*", |
| "*self_attn.q_b_proj*", |
| "*language_model.model.norm*", |
| "*language_model.model.output_attn_res_norm*", |
| "*language_model.model.output_attn_res_proj*", |
| "*mm_projector.post_norm*", |
| "*mlp.fc0*", |
| "*mlp.fc1*", |
| "*norm0*", |
| "*norm1*", |
| "*wo*", |
| "*wqkv*", |
| "*vision_tower.encoder.final_layernorm*", |
| "*vision_tower.patch_embed.pos_emb*", |
| "*vision_tower.patch_embed.proj*" |
| ] |
| }, |
| "remove_invalid_values": false, |
| "repetition_penalty": 1.0, |
| "return_dict": true, |
| "return_dict_in_generate": false, |
| "rms_norm_eps": 1e-05, |
| "routed_expert_hidden_size": 3584, |
| "routed_scaling_factor": 1.0, |
| "sep_token_id": null, |
| "suppress_tokens": null, |
| "task_specific_params": null, |
| "temperature": 1.0, |
| "tf_legacy_loss": false, |
| "tie_encoder_decoder": false, |
| "tie_word_embeddings": false, |
| "tokenizer_class": null, |
| "top_k": 50, |
| "top_p": 1.0, |
| "topk_group": 1, |
| "topk_method": "noaux_tc", |
| "torchscript": false, |
| "transformers_version": "4.56.2", |
| "typical_p": 1.0, |
| "use_bfloat16": false, |
| "use_cache": true, |
| "use_grouped_topk": true, |
| "v_head_dim": 128, |
| "vocab_size": 163840 |
| }, |
| "tie_word_embeddings": false, |
| "vision_config": { |
| "_attn_implementation": "flash_attention_2", |
| "activation_func": "gelu_pytorch_tanh", |
| "attn_bias": false, |
| "init_pos_emb_height": 64, |
| "init_pos_emb_time": 4, |
| "init_pos_emb_width": 64, |
| "linear_bias": false, |
| "merge_kernel_size": [ |
| 2, |
| 2 |
| ], |
| "merge_type": "sd2_tpool", |
| "mlp_type": "mlp2", |
| "mm_hidden_size": 1024, |
| "mm_projector_type": "patchmergerv2", |
| "norm_type": "rmsnorm", |
| "patch_embed_proj_bias": false, |
| "patch_size": 14, |
| "pos_emb_interpolation_mode": "bilinear", |
| "pos_emb_type": "divided_fixed", |
| "projector_hidden_act": "gelu", |
| "projector_ln_eps": 1e-05, |
| "qkv_hidden_size": 1536, |
| "text_hidden_size": 7168, |
| "vt_hidden_size": 1024, |
| "vt_intermediate_size": 4096, |
| "vt_num_attention_heads": 12, |
| "vt_num_hidden_layers": 27 |
| } |
| } |
