text_encoder/config.json

{
  "_class_name": "LTX2AudioVisualTextEncoder",
  "_diffusers_version": "0.36.0.dev0",
  "audio_connector_attention_head_dim": 8,
  "audio_connector_num_attention_heads": 4,
  "audio_connector_num_layers": 1,
  "audio_connector_num_learnable_registers": null,
  "causal_temporal_positioning": false,
  "config_only": true,
  "rope_base_seq_len": 32,
  "rope_double_precision": false,
  "rope_theta": 10000.0,
  "text_encoder_hidden_dim": 32,
  "text_model_id": "tiny-random/gemma-3",
  "text_proj_in_factor": 3,
  "video_connector_attention_head_dim": 8,
  "video_connector_num_attention_heads": 4,
  "video_connector_num_layers": 1,
  "video_connector_num_learnable_registers": null
}