MOSS-TTS-Nano is an open-source multilingual tiny speech generation model from MOSI.AI and the OpenMOSS team. With only 0.1B parameters, it is designed for realtime speech generation, can run directly on CPU without a GPU, and keeps the deployment stack simple enough for local demos, web serving, and lightweight product integration. 2026.4.10: We release MOSS-TTS-Nano. A demo Space is available at OpenMOSS-Team/MOSS-TTS-Nano. You can also view the demo and more details at openmoss.github.io/MOSS-TTS-Nano-Demo/. MOSS-TTS-Nano focuses on the part of TTS deployment that matters most in practice: small footprint, low latency, good enough quality for realtime products, and simple local setup. It…
Soprano-1.1-80M is an open-weight model for text to speech from Eugene Kwek, released under Apache License 2.0. It has 80M parameters and a 1,024-token context. At 16-bit it needs about 0.2 GB of GPU memory, which fits on 1x MI300X from $1.85 an hour, at the lowest prices in the SAVRN Index. It draws 52.9k downloads a month.
2026.01.14 - Soprano-1.1-80M released! 95% fewer hallucinations and a 63% preference rate over Soprano-80M. 2026.01.13 - Soprano-Factory released! You can now train/fine-tune your own Soprano models. 2025.12.22 - Soprano-80M released!
Parameters80M
Context1,024
Weights280.9 MB
Licenseapache-2.0
AccessOpen weights
Monthly Downloads52.9k
Runs On
What it takes to serve Soprano-1.1-80M (80M 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 | 0.2 GB | 0.2 GB | 1x MI300X (192 GB) Vultr |
$1.85 | 1x H100 $1.99 · 1x MI325X $2.00 |
| 8-bit | 0.1 GB | 0.1 GB | 1x MI300X (192 GB) Vultr |
$1.85 | 1x H100 $1.99 · 1x MI325X $2.00 |
| 4-bit | 0.0 GB | 0.0 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 Oct 16, 2026.
Soprano-1.1-80M on every accelerator the SAVRN Index prices, at every precision
Model Card
By Eugene Kwek, published under apache-2.0, revision 27b5a5f5f541.
Soprano: Instant, Ultra‑Realistic Text‑to‑Speech
[](https://github.com/ekwek1/soprano)
[](https://huggingface.co/spaces/ekwek/Soprano-TTS)
News
2026.01.14 - Soprano-1.1-80M released! 95% fewer hallucinations and a 63% preference rate over Soprano-80M.
2026.01.13 - Soprano-Factory released! You can now train/fine-tune your own Soprano models.
2025.12.22 - Soprano-80M released! Code | Demo
Overview
Soprano is an ultra‑lightweight, on-device text‑to‑speech (TTS) model designed for expressive, high‑fidelity speech synthesis at unprecedented speed. Soprano was designed with the following features: - Up to 2000x real-time generation on GPU and 20x real-time on CPU - Lossless streaming with <15 ms latency on GPU, <250 ms on CPU - <1 GB memory usage with a compact 80M parameter architecture - Infinite generation length with automatic text splitting - Highly expressive, crystal clear audio generation at 32kHz - Widespread support for CUDA, CPU, and MPS devices on Windows, Linux, and Mac - Supports WebUI, CLI, and OpenAI-compatible endpoint for easy and production-ready inference
Installation
Install with wheel (CUDA-only for now)
pip install soprano-tts
Configuration
- Architecture
- Qwen3ForCausalLM
- Context length (tokens)
- 1,024
- Layers
- 17
- Hidden size
- 512
- Feed-forward size
- 2,304
- Attention heads
- 4
- Key/value heads
- 1
- Head dimension
- 128
- Vocabulary size
- 8,192
- RoPE base
- 10,000
- Model type
- qwen3
Identity and Version
- Repository
- ekwek/Soprano-1.1-80M
- Publisher
- Eugene Kwek
- Task
- Text to speech
- Modality
- Audio
- Library
- transformers
- Parameters
- 80M parameters
- Languages
- Not stated by the source
- Revision
- 27b5a5f5f541a1db3a51d6fd1b0fc7147b92cd01
- First published
- 2026-01-14
- Last updated
- 2026-01-14
Files and Weights
9 files, 283.9 MB in total. The weights are 2 files totalling 280.9 MB in pth, safetensors.
Weights2 files · 280.9 MB
Configuration3 files · 1.3 KB
Tokenizer2 files · 3.0 MB
Documentation1 file · 6.6 KB
Repository1 file · 1.5 KB
Every file
| File | Type | Size | SHA-256 |
|---|---|---|---|
| decoder.pth | Weights | 121.4 MB | 11245487e458 |
| model.safetensors | Weights | 159.4 MB | 10e8c4223e14 |
| config.json | Configuration | 1.1 KB | — |
| generation_config.json | Configuration | 111 B | — |
| special_tokens_map.json | Configuration | 28 B | — |
| README.md | Documentation | 6.6 KB | — |
| .gitattributes | Repository | 1.5 KB | — |
| tokenizer.json | Tokenizer | 1.6 MB | — |
| tokenizer_config.json | Tokenizer | 1.4 MB | — |
License and Download
- License
- apache-2.0
- Access
- Open weights, no gate
- Download size
- 280.9 MB
Download from Eugene Kwek
Released by Eugene Kwek through its official repository on Hugging Face. Read the license.
Memory Requirements
| Precision | Weights in memory |
|---|---|
| As published | 280.9 MB |
| 16-bit | 0.2 GB |
| 8-bit | 0.1 GB |
| 4-bit | 0.0 GB |
Weights only, from the published parameter count; the key-value cache and runtime add to this.
Questions About Soprano-1.1-80M
How much GPU memory does Soprano-1.1-80M need?
About 0.2 GB at 16-bit and 0 GB at 4-bit: the weights (80M parameters) plus a working margin. A long context needs more.
What is the cheapest GPU to run Soprano-1.1-80M 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 Soprano-1.1-80M commercially?
Yes. Soprano-1.1-80M 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.
What is Soprano-1.1-80M's context length?
1,024 tokens, from the maximum position embeddings in its published configuration.
Similar Models
mms - vits pipelinetag: text-to-speech This repository contains the Hindi (hin) language text-to-speech (TTS) model checkpoint. This model is part of Facebook's Massively Multilingual Speech project, aiming to provide speech technology across a diverse range of languages. You can find more details about the supported languages and their ISO 639-3 codes in the MMS Language Coverage Overview, and see all MMS-TTS checkpoints on the Hugging Face Hub: facebook/mms-tts. MMS-TTS is available in the Transformers library from version 4.33 onwards. VITS (Variational Inference with adversarial learning for end-to-end Text-to-Speech) is an end-to-end speech synthesis model that predicts a speech…
mms - vits pipelinetag: text-to-speech This repository contains the English (eng) language text-to-speech (TTS) model checkpoint. This model is part of Facebook's Massively Multilingual Speech project, aiming to provide speech technology across a diverse range of languages. You can find more details about the supported languages and their ISO 639-3 codes in the MMS Language Coverage Overview, and see all MMS-TTS checkpoints on the Hugging Face Hub: facebook/mms-tts. MMS-TTS is available in the Transformers library from version 4.33 onwards. VITS (Variational Inference with adversarial learning for end-to-end Text-to-Speech) is an end-to-end speech synthesis model that predicts a speech…
mms - vits pipelinetag: text-to-speech This repository contains the Bamanankan (bam) language text-to-speech (TTS) model checkpoint. This model is part of Facebook's Massively Multilingual Speech project, aiming to provide speech technology across a diverse range of languages. You can find more details about the supported languages and their ISO 639-3 codes in the MMS Language Coverage Overview, and see all MMS-TTS checkpoints on the Hugging Face Hub: facebook/mms-tts. MMS-TTS is available in the Transformers library from version 4.33 onwards. VITS (Variational Inference with adversarial learning for end-to-end Text-to-Speech) is an end-to-end speech synthesis model that predicts a speech…
VieNeu-TTS v3 Turbo is the next generation of Vietnamese TTS — 48 kHz high-fidelity speech, 23 built-in preset voices across three regions (North / Central / South), instant voice cloning, real-time streaming with an OpenAI-compatible API (16 concurrent streams on one RTX 3060), inline emotion cues, and seamless bilingual (En–Vi) code-switching. The reference implementation is the vieneu Python SDK (v3.7.1). Its minimal install is torch-free: on CPU everything runs on ONNX Runtime (PyTorch is never imported), and on a CUDA machine it auto-switches to the PyTorch engine with automatic batching and a continuous-batching stream scheduler — same API, no code change. The VieNeu-TTS v3 Turbo…
Zero-shot voice tuner for Kokoro-82M. Outputs base Kokoro compatible voice packs @ [510, 1, 256]. (write-up) Same passage for every voice, enrolled from the references. LibriTTS-R speakers are dev-clean held out from training. Integrated into Kokoro-FastAPI (v0.9.0+) The pack is a plain tensor; torch.save(pack, "voices/amme.pt") makes it a voice file like any other, prefixed by accent and gender like the stock packs. The pitch-tracking ceiling is set automatically from the reference's harmonic spacing, so band-limited or archival sources land in the right octave without tuning. - enroll(..., fmax=180) overrides it if a voice still reads the wrong register. Enrollment embeds an input audio…