{ "architectures" : [ "DiffusionGemmaForBlockDiffusion" ], "boi_token_id" : 255999, "canvas_length" : 256, "dtype" : "bfloat16", "eoi_token_id" : 258882, "eos_token_id" : [ 1, 106 ], "image_token_id" : 258880, "initializer_range" : 0.02, "model_type" : "diffusion_gemma", "quantization" : { "bits" : 4, "group_size" : 64, "mode" : "affine" }, "quantization_config" : { "bits" : 4, "group_size" : 64, "mode" : "affine" }, "text_config" : { "attention_bias" : 0, "attention_dropout" : 0, "bos_token_id" : 2, "dtype" : "bfloat16", "eos_token_id" : 1, "final_logit_softcapping" : 30, "global_head_dim" : 512, "head_dim" : 256, "hidden_activation" : "gelu_pytorch_tanh", "hidden_size" : 2816, "initializer_range" : 0.02, "intermediate_size" : 2112, "layer_types" : [ "sliding_attention", "sliding_attention", "sliding_attention", "sliding_attention", "sliding_attention", "full_attention", "sliding_attention", "sliding_attention", "sliding_attention", "sliding_attention", "sliding_attention", "full_attention", "sliding_attention", "sliding_attention", "sliding_attention", "sliding_attention", "sliding_attention", "full_attention", "sliding_attention", "sliding_attention", "sliding_attention", "sliding_attention", "sliding_attention", "full_attention", "sliding_attention", "sliding_attention", "sliding_attention", "sliding_attention", "sliding_attention", "full_attention" ], "max_position_embeddings" : 262144, "model_type" : "diffusion_gemma_text", "moe_intermediate_size" : 704, "num_attention_heads" : 16, "num_experts" : 128, "num_global_key_value_heads" : 2, "num_hidden_layers" : 30, "num_key_value_heads" : 8, "pad_token_id" : 0, "rms_norm_eps" : 9.9999999999999995e-07, "rope_parameters" : { "full_attention" : { "partial_rotary_factor" : 0.25, "rope_theta" : 1000000, "rope_type" : "proportional" }, "sliding_attention" : { "rope_theta" : 10000, "rope_type" : "default" } }, "sliding_window" : 1024, "tie_word_embeddings" : 1, "top_k_experts" : 8, "use_bidirectional_attention" : "vision", "vocab_size" : 262144 }, "tie_word_embeddings" : 1, "transformers_version" : "5.8.0.dev0", "vision_config" : { "_name_or_path" : "", "architectures" : null, "attention_bias" : 0, "attention_dropout" : 0, "chunk_size_feed_forward" : 0, "default_output_length" : 280, "dtype" : "bfloat16", "global_head_dim" : 72, "head_dim" : 72, "hidden_activation" : "gelu_pytorch_tanh", "hidden_size" : 1152, "id2label" : { "0" : "LABEL_0", "1" : "LABEL_1" }, "initializer_range" : 0.02, "intermediate_size" : 4304, "is_encoder_decoder" : 0, "label2id" : { "LABEL_0" : 0, "LABEL_1" : 1 }, "max_position_embeddings" : 131072, "model_type" : "gemma4_vision", "num_attention_heads" : 16, "num_hidden_layers" : 27, "num_key_value_heads" : 16, "output_attentions" : 0, "output_hidden_states" : 0, "patch_size" : 16, "pooling_kernel_size" : 3, "position_embedding_size" : 10240, "problem_type" : null, "return_dict" : 1, "rms_norm_eps" : 9.9999999999999995e-07, "rope_parameters" : { "rope_theta" : 100, "rope_type" : "default" }, "standardize" : 1, "use_clipped_linears" : 0 }, "vision_soft_tokens_per_image" : 280 }