Qwen3-TTS covers 10 major languages (Chinese, English, Japanese, Korean, German, French, Russian, Portuguese, Spanish, and Italian) as well as multiple dialectal voice profiles to meet global application needs.
Runs On
What it takes to serve Qwen3-TTS-12Hz-1.7B-Base (1.9B parameters): the memory its weights need at each precision, and the cheapest way to rent enough data-center GPUs to hold them.
| Precision | Weights | Memory needed | Cheapest setup | Per hour | Also fits |
|---|---|---|---|---|---|
| 16-bit | 3.9 GB | 4.6 GB | 1x MI300X (192 GB) Vultr |
$1.85 | 1x H100 $1.99 · 1x MI325X $2.00 |
| 8-bit | 1.9 GB | 2.3 GB | 1x MI300X (192 GB) Vultr |
$1.85 | 1x H100 $1.99 · 1x MI325X $2.00 |
| 4-bit | 1.0 GB | 1.2 GB | 1x MI300X (192 GB) Vultr |
$1.85 | 1x H100 $1.99 · 1x MI325X $2.00 |
Memory is the weights at that precision plus 20% for the runtime and a short context; a long context needs more. Prices are the lowest on-demand hourly rates in the SAVRN Index, read Sep 18, 2026.
SAVRN's Notes on Qwen3-TTS-12Hz-1.7B-Base
The name says 1.7B and the configuration counts 1,928,677,440 parameters, so budget for 1.9B. That puts 16-bit memory at 4.6 GB for 3.9 GB of weights and 8-bit at 2.3 GB, both a sliver of the cheapest setup we list, one MI300X with 192 GB at $1.85 an hour. Its purpose is speech generation across ten languages, Chinese, English, Japanese, Korean, German, French, Russian, Portuguese, Spanish and Italian, with dialect voice profiles and control of tone, speaking rate and emotion from natural-language instructions.
Deploying it commercially is allowed; Apache 2.0 covers modification and redistribution too, as long as the license and NOTICE file stay and significant changes are stated. Check the empty library field. Nothing in the record says which serving stack loads Qwen3TTSForConditionalGeneration, so prove that before you buy time. Then read the technical report, arXiv:2601.15621, for what the Base suffix means against the instruction-driven control described here.
Model Card
By Qwen, published under apache-2.0, revision fd4b25438912.
Qwen3-TTS
Overview
Introduction
Configuration
- Architecture
- Qwen3TTSForConditionalGeneration
- Model type
- qwen3_tts
Identity and Version
- Repository
- Qwen/Qwen3-TTS-12Hz-1.7B-Base
- Publisher
- Qwen
- Task
- Not stated by the source
- Modality
- Other
- Library
- Not stated by the source
- Parameters
- 1.9B parameters
- Languages
- Not stated by the source
- Revision
- fd4b254389122332181a7c3db7f27e918eec64e3
- First published
- 2026-01-21
- Last updated
- 2026-01-23
Files and Weights
13 files, 4.5 GB in total. The weights are 2 files totalling 4.5 GB in safetensors.
Every file
| File | Type | Size | SHA-256 |
|---|---|---|---|
| model.safetensors | Weights | 3.9 GB | 38fc7fc51c5e |
| speech_tokenizer/model.safetensors | Weights | 682.3 MB | 836b7b357f5e |
| config.json | Configuration | 4.5 KB | — |
| generation_config.json | Configuration | 245 B | — |
| preprocessor_config.json | Configuration | 127 B | — |
| speech_tokenizer/config.json | Configuration | 2.3 KB | — |
| speech_tokenizer/configuration.json | Configuration | 76 B | — |
| speech_tokenizer/preprocessor_config.json | Configuration | 234 B | — |
| README.md | Documentation | 57.8 KB | — |
| .gitattributes | Repository | 1.5 KB | — |
| merges.txt | Tokenizer | 1.7 MB | — |
| tokenizer_config.json | Tokenizer | 7.3 KB | — |
| vocab.json | Tokenizer | 2.8 MB | — |
License and Download
- License
- apache-2.0
- Access
- Open weights, no gate
- Download size
- 4.5 GB
Released by Qwen through ModelScope. Read the license.
Built From
- Described by arXiv:2601.15621
Memory Requirements
| Precision | Weights in memory |
|---|---|
| As published | 4.5 GB |
| 16-bit | 3.9 GB |
| 8-bit | 1.9 GB |
| 4-bit | 1.0 GB |
Weights only, from the published parameter count; the key-value cache and runtime add to this.
Built on This Model
- Adapter ofQwen3-Voice-Embedding-12Hz-1.7B
- Derived fromQwen3-Voice-Embedding-12Hz-1.7B
Questions About Qwen3-TTS-12Hz-1.7B-Base
How much GPU memory does Qwen3-TTS-12Hz-1.7B-Base need?
About 4.6 GB at 16-bit and 1.2 GB at 4-bit: the weights (1.9B parameters) plus a working margin. A long context needs more.
What is the cheapest GPU to run Qwen3-TTS-12Hz-1.7B-Base on?
At 16-bit, 1x MI300X from $1.85 an hour; at 4-bit, 1x MI300X from $1.85 an hour, at the lowest on-demand prices the SAVRN Index lists.
Can I use Qwen3-TTS-12Hz-1.7B-Base commercially?
Yes. Qwen3-TTS-12Hz-1.7B-Base is released under Apache License 2.0. The Apache License 2.0 is a permissive open-source license. It permits commercial use, modification and redistribution. It requires keeping the license and copyright notices and any NOTICE file, stating significant changes, and it includes an express patent grant from contributors.