{ "model_type": "speech-llm", "features_size": 128, "preprocessor": "whisper128", "hidden_size": 3072, "num_layers": 30, "num_key_value_heads": 8, "head_dim": 128, "eos_token_ids": [ 2 ], "max_sequence_length": 512, "prompt_prefix_ids": [ 1, 3, 25 ], "prompt_suffix_ids": [ 4, 34 ], "language_suffix_ids": { "en": [ 4, 9909, 1058, 1262, 34 ], "English": [ 4, 9909, 1058, 1262, 34 ], "fr": [ 4, 9909, 1058, 7064, 34 ], "French": [ 4, 9909, 1058, 7064, 34 ], "de": [ 4, 9909, 1058, 1558, 34 ], "German": [ 4, 9909, 1058, 1558, 34 ], "es": [ 4, 9909, 1058, 1264, 34 ], "Spanish": [ 4, 9909, 1058, 1264, 34 ], "it": [ 4, 9909, 1058, 1276, 34 ], "Italian": [ 4, 9909, 1058, 1276, 34 ], "pt": [ 4, 9909, 1058, 1515, 34 ], "Portuguese": [ 4, 9909, 1058, 1515, 34 ], "nl": [ 4, 9909, 24082, 1108, 34 ], "Dutch": [ 4, 9909, 24082, 1108, 34 ], "hi": [ 4, 9909, 1058, 8101, 34 ], "Hindi": [ 4, 9909, 1058, 8101, 34 ] }, "tokenizer_type": "byte-level", "source_model": "mistralai/Voxtral-Mini-3B-2507" }