{ "architectures": [ "Glm5NextVisionModel" ], "attention_bias": true, "attention_dropout": 0.0, "depth": 2, "dtype": "float32", "hidden_act": "silu", "hidden_size": 48, "image_size": 336, "in_channels": 3, "initializer_range": 0.02, "intermediate_size": 16, "model_type": "glm5_next_vision", "num_heads": 2, "out_hidden_size": 48, "patch_size": 14, "projection_intermediate_size": 144, "rms_norm_eps": 1e-05, "rope_parameters": { "rope_theta": 10000.0, "rope_type": "axial" }, "spatial_merge_size": 1, "swiglu_limit": 10.0, "temporal_patch_size": 2, "transformers_version": "5.19.0.dev0" }