{ "architectures": [ "UniARForConditionalGeneration" ], "model_type": "uniar", "text_config": { "vocab_size": 151936, "hidden_size": 4096, "intermediate_size": 12288, "num_hidden_layers": 36, "num_attention_heads": 32, "num_key_value_heads": 8, "head_dim": 128, "hidden_act": "silu", "max_position_embeddings": 32768, "initializer_range": 0.02, "rms_norm_eps": 1e-06, "use_cache": true, "rope_theta": 1000000, "rope_scaling": { "mrope_section": [ 24, 20, 20 ], "rope_type": "default" }, "attention_bias": false, "attention_dropout": 0.0, "tie_word_embeddings": false }, "vision_config": { "depth": 27, "hidden_size": 1152, "hidden_act": "gelu_pytorch_tanh", "intermediate_size": 4304, "num_heads": 16, "in_channels": 3, "patch_size": 16, "spatial_merge_size": 2, "temporal_patch_size": 2, "out_hidden_size": 4096, "initializer_range": 0.02, "use_bsq": true, "bsq_dim": 64, "bsq_hidden_dim": 8192, "bsq_skip_final_layernorm": true, "vistok_pred": false, "vistok_pred_layernorm": false, "vistok_pred_transformer_head": false, "num_position_embeddings": 2304, "deepstack_visual_indexes": [ 8, 16, 24 ] }, "vistok_pred": true, "vistok_pred_layernorm": true, "vistok_pred_use_mlp": false, "visual_transformer_decoder": true, "image_token_id": 151655, "video_token_id": 151656, "vision_start_token_id": 151652, "vision_end_token_id": 151653, "tie_word_embeddings": false, "torch_dtype": "bfloat16" }