aitytech/Qwen3-TTS-12Hz-0.6B-Base-MLX-4bit

Qwen3 TTS 12Hz 0.6B Base — MLX 4-bit

MLX 4-bit quantized conversion of Qwen/Qwen3-TTS-12Hz-0.6B-Base for Apple Silicon inference.

Usage

Used by qwen3-asr-swift Qwen3TTS module:

let model = try await Qwen3TTSModel.fromPretrained()
let audio = try model.synthesize("Hello, world!")

audio speak "Hello, world!" -o output.wav

Model Details

• Architecture: Qwen3-TTS (Talker transformer + Code Predictor + Mimi speech tokenizer decoder)
• Parameters: 0.6B
• Quantization: 4-bit (MLX, talker only)
• Size: ~981 MB
• Sample rate: 24 kHz
• Codec rate: 12.5 Hz