config.json

{
  "architectures": [
    "HiggsAudioV2ForConditionalGeneration"
  ],
  "attention_bias": false,
  "attention_dropout": 0.0,
  "audio_bos_token_id": 128013,
  "audio_delay_token_id": 128014,
  "audio_stream_bos_id": 1024,
  "audio_stream_eos_id": 1025,
  "audio_token_id": 128016,
  "bos_token_id": 1,
  "codebook_size": 1026,
  "dtype": "bfloat16",
  "eos_token_id": 128009,
  "head_dim": 128,
  "hidden_act": "silu",
  "hidden_size": 3072,
  "initializer_range": 0.02,
  "intermediate_size": 8192,
  "max_position_embeddings": 2048,
  "mlp_bias": false,
  "model_type": "higgs_audio_v2",
  "num_attention_heads": 24,
  "num_codebooks": 8,
  "num_hidden_layers": 28,
  "num_key_value_heads": 8,
  "pad_token_id": 128001,
  "pretraining_tp": 1,
  "rms_norm_eps": 1e-05,
  "rope_parameters": {
    "factor": 32.0,
    "high_freq_factor": 0.5,
    "low_freq_factor": 0.125,
    "original_max_position_embeddings": 1024,
    "rope_theta": 500000.0,
    "rope_type": "llama3"
  },
  "tie_word_embeddings": false,
  "transformers_version": "5.3.0.dev0",
  "use_cache": true,
  "vocab_size": 128256
}