diffusers/LTX-2.3-Diffusers
41.2k
1{2 "_class_name": "LTX2VideoTransformer3DModel",3 "_diffusers_version": "0.37.0.dev0",4 "activation_fn": "gelu-approximate",5 "attention_bias": true,6 "attention_head_dim": 128,7 "attention_out_bias": true,8 "audio_attention_head_dim": 64,9 "audio_cross_attention_dim": 2048,10 "audio_cross_attn_mod": true,11 "audio_gated_attn": true,12 "audio_hop_length": 160,13 "audio_in_channels": 128,14 "audio_num_attention_heads": 32,15 "audio_out_channels": 128,16 "audio_patch_size": 1,17 "audio_patch_size_t": 1,18 "audio_pos_embed_max_pos": 20,19 "audio_sampling_rate": 16000,20 "audio_scale_factor": 4,21 "base_height": 2048,22 "base_width": 2048,23 "caption_channels": 3840,24 "causal_offset": 1,25 "cross_attention_dim": 4096,26 "cross_attn_mod": true,27 "cross_attn_timestep_scale_multiplier": 1000,28 "gated_attn": true,29 "in_channels": 128,30 "norm_elementwise_affine": false,31 "norm_eps": 1e-06,32 "num_attention_heads": 32,33 "num_layers": 48,34 "out_channels": 128,35 "patch_size": 1,36 "patch_size_t": 1,37 "perturbed_attn": true,38 "pos_embed_max_pos": 20,39 "qk_norm": "rms_norm_across_heads",40 "rope_double_precision": true,41 "rope_theta": 10000.0,42 "rope_type": "split",43 "timestep_scale_multiplier": 1000,44 "use_prompt_embeddings": false,45 "vae_scale_factors": [46 8,47 32,48 3249 ]50}51 