Serveurperso / text-to-speech updated 1 week ago

Qwen3-TTS-GGUF

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.

Params
Context
Downloads 30d
395 K
Likes
45
Commercial use: allowed apache-2.0 Not gated GGUF 11 languages View on Hugging Face ↗

Download history

daily snapshots · 39 days
▲ 395 K in the last 30 days (50.0%)
901 K394 K
Aug 14Aug 27Sep 9Sep 21

Can you run it?

Estimated VRAM at 8K context unless noted. Pick your hardware to see the verdict per quantization.

FileQuantSizeEst. VRAMVerdict on RTX 4090 · 24 GB
qwen-talker-1.7b-base-Q4_K_M.gguf Q4_K_M 1.2 GB 1.8 GB ✅ Runs comfortably
qwen-talker-1.7b-base-Q8_0.gguf Q8_0 2.1 GB 2.8 GB ✅ Runs comfortably
qwen-talker-1.7b-base-F32.gguf GGUF 7.7 GB 9.0 GB ✅ Runs comfortably
Estimate: file size × 1.1 + KV cache at 8K + 0.5 GB overhead. Not a benchmark — how we calculate this.

Specifications

Licence
apache-2.0
First seen on the Hub
2026-05-11
Training datasets
undisclosed
Added to our catalog
2026-08-14
Compare with any text-to-speech model