aufklarer/Qwen3-TTS-12Hz-1.7B-Base-MLX-4bit

Qwen3 TTS 12Hz 1.7B Base — MLX 4-bit

MLX 4-bit quantized conversion of Qwen/Qwen3-TTS-12Hz-1.7B-Base for Apple Silicon inference.

Usage

Used by speech-swift Qwen3TTS module:

let model = try await Qwen3TTSModel.fromPretrained(
    modelId: "aufklarer/Qwen3-TTS-12Hz-1.7B-Base-MLX-4bit"
)
let audio = try model.synthesize("Hello, world!")

audio speak "Hello, world!" --model 1.7b -o output.wav

Model Details

• Architecture: Qwen3-TTS (Talker transformer + Code Predictor + Mimi speech tokenizer decoder)
• Parameters: 1.7B
• Quantization: 4-bit (MLX, talker + code predictor)
• Size: ~1.7 GB
• Sample rate: 24 kHz
• Codec rate: 12.5 Hz

Variants

Variant	Quantization	Size	Model ID
0.6B 4-bit	4-bit	~981 MB	aufklarer/Qwen3-TTS-12Hz-0.6B-Base-MLX-4bit
0.6B 8-bit	8-bit	~1.3 GB	aufklarer/Qwen3-TTS-12Hz-0.6B-Base-MLX-8bit
1.7B 4-bit	4-bit	~1.7 GB	aufklarer/Qwen3-TTS-12Hz-1.7B-Base-MLX-4bit
1.7B 8-bit	8-bit	~2.8 GB	aufklarer/Qwen3-TTS-12Hz-1.7B-Base-MLX-8bit

---

---

• Guide: soniqo.audio/guides/speak
• Docs: soniqo.audio
• GitHub: soniqo/speech-swift