Have a ComfyUI workflow? Get a readable Workflow Doc in your browser.Make a workflow readable.Try it now ↗
GGUF weights for qwentts.cpp, a C++17/GGML port of Qwen3-TTS 12 Hz (Qwen team, Alibaba). Multilingual zero shot TTS with named speakers and Mandarin dialects, 24 kHz mono. Runs on CPU, CUDA, Metal, Vulkan.
Model source
Source description
GGUF weights for qwentts.cpp, a C++17/GGML port of Qwen3-TTS 12 Hz (Qwen team, Alibaba). Multilingual zero shot TTS with named speakers and Mandarin dialects, 24 kHz mono. Runs on CPU, CUDA, Metal, Vulkan.
Two GGUFs load together :
qwen-talker-{size}-{mode}-{variant}.gguf Qwen3 LM + code predictor MTP head + optional speaker encoder, text -> 12 Hz codes qwen-tokenizer-12hz-{variant}.gguf SEANet + ConvNeXt + DAC v2 + RVQ, 12 Hz codes <-> 24 kHz audio
Sources
1 sourceVerified Aug 7
Model artifacts
24 artifactsSource excerpts
2 excerptsThree modes are available across two talker sizes :
| mode | 0.6B | 1.7B | use case |
|---|---|---|---|
| base | yes | yes | zero shot TTS with named speakers and dialects |
| customvoice | yes | yes | zero shot voice cloning from a reference clip |
| voicedesign | no | yes | voice synthesis from attribute description |
The tokenizer is shared across every talker.
| variant | talker 0.6B | talker 1.7B | tokenizer | use case |
|---|---|---|---|---|
| F32 | 3.7 GB | 7.7 GB | 647 MB | reference, debug, conversion |
| BF16 | 1.8 GB | 3.9 GB | 359 MB | source faithful, max precision |
| Q8_0 | 993 MB | 2.1 GB | 291 MB | recommended default |
| Q4_K_M | 629 MB | 1.2 GB | 255 MB | lowest VRAM |
git clone --recurse-submodules https://github.com/ServeurpersoCom/qwentts.cpp.git
cd qwentts.cpp && ./buildcuda.sh
mkdir -p models
huggingface-cli download Serveurperso/Qwen3-TTS-GGUF \
qwen-talker-1.7b-base-Q8_0.gguf qwen-tokenizer-12hz-Q8_0.gguf \
--local-dir models
cd examples
./base.sh # named speaker -> base.wav
./clone.sh # voice cloning -> clone.wav
./customvoice.sh # custom voice mode -> customvoice.wav
./tts.sh # voice design -> tts.wav
Set GGML_BACKEND to force a device, otherwise the runtime picks the
best one available.
| value | target |
|---|---|
CUDA0 | NVIDIA GPU, fastest path on Ada / Blackwell |
Vulkan0 | Cross vendor GPU (AMD / Intel / NVIDIA) |
Metal | Apple Silicon GPU |
CPU | CPU fallback, x86 variant auto selected |
Tokenizer GGUFs are not uniform quants. Three categories get a dedicated treatment :
| tensor | dtype across all variants |
|---|---|
| RVQ codebooks, input_proj / output_proj, speaker encoder fc | F32 |
| 1D tensors (gamma, biases, norms, snake alpha and beta) | F32 |
| Conv kernels with non alignable rows (K=7,3,1) | F16 in Q* variants |
Conv kernel rows (K=7,3,1) never divide a K-quant block size, so the
quantizer skips the Q* intermediates and lands on F16 directly. This
is the last resort branch of llama.cpp's tensor_type_fallback
applied unconditionally for these kernels. F16 has no block size and
matches the runtime target dtype on every backend. The talker LM
(Qwen3 backbone, hidden divisible by 256) follows standard llama.cpp
K-quant across variants. The code predictor MTP head and the speaker
encoder live in the talker GGUF and share its quantization.
Upstream model : Qwen3-TTS by Alibaba / Qwen team, Apache 2.0 Audio codec : Qwen3-TTS-Tokenizer-12Hz (Qwen team), Apache 2.0 GGUF tooling : qwentts.cpp, MIT
qwen-talker-0.6b-base-Q4_K_M.gguf
gguf · 600 MB · SHA-256 4b468ec7b1f6…4226 · Hugging Face
qwen-talker-0.6b-customvoice-BF16.gguf
gguf · 1.69 GB · SHA-256 389a153f03e1…8d3e · Hugging Face
Downloadqwen-talker-0.6b-customvoice-F32.gguf
gguf · 3.38 GB · SHA-256 436d1f130ee8…4f37 · Hugging Face
Downloadqwen-talker-0.6b-customvoice-Q4_K_M.gguf
gguf · 577 MB · SHA-256 b3a7e6613d80…8e1e · Hugging Face
Downloadqwen-talker-0.6b-customvoice-Q8_0.gguf
gguf · 924 MB · SHA-256 4eb38675c736…8519 · Hugging Face
Downloadqwen-talker-1.7b-customvoice-BF16.gguf
gguf · 3.58 GB · SHA-256 dd5d70d3f7d0…276c · Hugging Face
Downloadqwen-talker-1.7b-customvoice-F32.gguf
gguf · 7.15 GB · SHA-256 f7df5b2be1d8…251b · Hugging Face
Downloadqwen-talker-1.7b-customvoice-Q4_K_M.gguf
gguf · 1.10 GB · SHA-256 cc328834a631…fc4a · Hugging Face
Downloadqwen-talker-1.7b-customvoice-Q8_0.gguf
gguf · 1.90 GB · SHA-256 cab2cff67a0a…89fa · Hugging Face
Downloadqwen-talker-1.7b-voicedesign-BF16.gguf
gguf · 3.58 GB · SHA-256 f389939fed3c…ad42 · Hugging Face
Downloadqwen-talker-1.7b-voicedesign-F32.gguf
gguf · 7.15 GB · SHA-256 b32c8571977f…ee90 · Hugging Face
Downloadqwen-talker-1.7b-voicedesign-Q4_K_M.gguf
gguf · 1.10 GB · SHA-256 7605ed0cc5e7…eee3 · Hugging Face
Downloadqwen-talker-1.7b-voicedesign-Q8_0.gguf
gguf · 1.90 GB · SHA-256 575610ab1ddc…2276 · Hugging Face
Download--- license: apache-2.0 library_name: gguf pipeline_tag: text-to-speech tags: - tts - text-to-speech - voice-cloning - voice-design - mandarin-dialects - ggml - gguf - qwen - qwen3 - qwen3-tts - cpp language: - zh - en - fr - de - es - it - pt - ja - ko - ru - ar base_model: - Qwen/Qwen3-TTS-Tokenizer-12Hz - Qwen/Qwen3-TTS-12Hz-0.6B-Base - Qwen/Qwen3-TTS-12Hz-0.6B-CustomVoice - Qwen/Qwen3-TTS-12Hz-1.7B-Base - Qwen/Qwen3-TTS-12Hz-1.7B-CustomVoice - Qwen/Qwen3-TTS-12Hz-1.7B-VoiceDesign --- # Qwen3-TTS GGUF GGUF weights for [qwentts.cpp](https://github.com/ServeurpersoCom/qwentts.cpp), a C++17/GGML port of Qwen3-TTS 12 Hz (Qwen team, Alibaba). Multilingual zero shot TTS with named speakers and Mandarin dialects, 24 kHz mono. Runs on CPU, CUDA, Metal, Vulkan. ## Files Two GGUFs load together : qwen-talker-{size}-{mode}-{variant}.gguf Qwen3 LM + code predictor MTP head + optional speaker encoder, text -> 12 Hz codes qwen-tokenizer-12hz-{variant}.gguf SEANet + ConvNeXt + DAC v2 + RVQ, 12 Hz codes <-> 24 kHz audio Three modes are available across two talker sizes : | mode | 0.6B | 1.7B | use case | |-------------|------|------|---------------------------------------------------| | base | yes | yes | zero shot TTS with named speakers and dialects | | customvoice | yes | yes | zero shot voice cloning from a reference clip | | voicedesign | no | yes | voice synthesis from attribute description | The tokenizer is shared across every talker. | variant | talker 0.6B | talker 1.7B | tokenizer | use case | |---------|-------------|-------------|-----------|--------------------------------| | F32 | 3.7 GB | 7.7 GB | 647 MB | reference, debug, conversion | | BF16 | 1.8 GB | 3.9 GB | 359 MB | source faithful, max precision | | Q8_0 | 993 MB | 2.1 GB | 291 MB | recommended default | | Q4_K_M | 629 MB | 1.2 GB | 255 MB | lowest VRAM | ## Quick start ``` git clone --recurse-submodules https://github.com/ServeurpersoCom/qwentts.cpp.git cd qwentts.cpp && ./buildcuda.sh mkdir -p models huggingface-cli download Serveurperso/Qwen3-TTS-GGUF \ qwen-talker-1.7b-base-Q8_0.gguf qwen-tokenizer-12hz-Q8_0.gguf \ --local-dir models cd examples ./base.sh # named speaker -> base.wav ./clone.sh # voice cloning -> clone.wav ./customvoice.sh # custom voice mode -> customvoice.wav ./tts.sh # voice design -> tts.wav ``` ## Backends Set `GGML_BACKEND` t...
Source context: 1332 downloads · 0 likes · Pipeline text-to-speech · Library gguf · Repo CC-TM/Qwen3-TTS-GGUF