Model Intelligence Sheet
ybalakhonov/qwen3-tts-gguf overview
Qwen3 TTS GGUF for qwen3 tts.cpp GGUF weights for on device Qwen3 TTS, used by the FoxGPT desktop app's local TTS. Source model: Qwen/Qwen3 TTS 12Hz 0.6B Base …
Runs locally from ~325.4 MB disk (4 GB VRAM class GPUs with llama.cpp / guIDE).
Repository Files & Downloads
Model Details
| Model ID | ybalakhonov/qwen3-tts-gguf |
|---|---|
| Author | ybalakhonov |
| Pipeline | — |
| License | apache-2.0 |
| Base model | — |
| Last modified | 2026-07-02T15:54:29.000Z |
Model README
---
license: apache-2.0
tags: [tts, qwen3-tts, gguf, foxgpt]
---
Qwen3-TTS GGUF (for qwen3-tts.cpp)
GGUF weights for on-device Qwen3-TTS, used by the FoxGPT desktop app's local TTS.
- Source model: Qwen/Qwen3-TTS-12Hz-0.6B-Base
- Converted with: predict-woo/qwen3-tts.cpp
setup_pipeline_models.py - Compatible ONLY with predict-woo/qwen3-tts.cpp (custom tensor layout; NOT llama.cpp TTS GGUF).
Layout
qwen3-tts-12hz-0.6b/
qwen3-tts-0.6b-f16.gguf # transformer + text tokenizer (~1.84 GB)
qwen3-tts-tokenizer-f16.gguf # vocoder/audio tokenizer (~341 MB)Point qwen3-tts-cli -m <dir> at a folder containing both files.
Run ybalakhonov/qwen3-tts-gguf with guIDE
Download guIDE — the AI-native code editor with local LLM inference and 69 built-in tools.
Source: Hugging Face · Compare models