Gated, non-commercial (CC BY-NC 4.0): ask for access on this page. Cilla Mini is the 18M model for CPUs, phones and browsers: Cilla shrunk with Kyutai's small-model recipe (8 layers, a 64-dim interface, a one-step drifting sampler head, the 2.6M Mimi decoder), guidance baked in.
Alternatives to cilla-mini
120 open-weight models that do the same job as cilla-mini (28M TTS), ranked by how close they are in task and size, then by popularity and how recent they are. Licence and VRAM are shown for each.
Soprano is an ultra‑lightweight, on-device text‑to‑speech (TTS) model designed for expressive, high‑fidelity speech synthesis at unprecedented speed.
80M512 ctx Apache 2.0182 314Dec 2025- Same category
- Larger (80M vs 28M)
- Commercial use OK
If TeraTTSv2 is useful to you, consider supporting its development. Thank you for your support!
?Bterattsv2-openrail-m-commercial-attribution4K 38Aug 2026- Same category
OmniVoice is a massively multilingual zero-shot text-to-speech (TTS) model supporting over 600 languages. Built on a novel diffusion language model-style architecture, it delivers high-quality speech with superior inference speed, supporting voice cloning and voice design.
613MUnknown licence4.2M 1.7KMar 2026- Same category
- Larger (613M vs 28M)
- Widely used
tiny-voca is a small English text-to-speech model with two voices, F2 (female) and M2 (male). The complete model, including the waveform generator, has 4.5M parameters and was trained from scratch on about 18 hours of synthetic speech. It runs about 70x faster than real time on four CPU threads.
5MApache 2.06.4K 338Nov 2026- Same category
- Smaller (5M vs 28M)
- Commercial use OK
A complete voice. A smaller footprint. Six English voices. Local generation. More ways to make speech your own.
?BApache 2.0805 47Nov 2026- Same category
- Commercial use OK
AuK is a 1.5B foundation model for speech generation and editing. Trained on millions of hours of diverse audio data, AuK supports zero-shot and instruction-based TTS, content and acoustic editing, paralinguistic editing, speech enhancement, and source separation through a unified natural-language…
?BMIT1.4K 396Aug 2026- Same category
- Commercial use OK
IndexTTS2: A Breakthrough in Emotionally Expressive and Duration-Controlled Auto-Regressive Zero-Shot Text-to-Speech
?BUnknown licence26K 796Jun 2025- Same category
VoxCPM2 is a tokenizer-free, diffusion autoregressive Text-to-Speech model — 2B parameters, 30 languages, 48kHz audio output, trained on over 2 million hours of multilingual speech data.
2.3BApache 2.0971.2K 1.8KApr 2026- Same category
- Larger (2.3B vs 28M)
- Commercial use OK
AuK is a 1.5B foundation model for speech generation and editing. Trained on millions of hours of diverse audio data, AuK supports zero-shot and instruction-based TTS, content and acoustic editing, paralinguistic editing, speech enhancement, and source separation through a unified natural-language…
?BMIT664 107Aug 2026- Same category
- Commercial use OK
E2-TTS is an open-weight tts model from SWivid, published on Hugging Face.
?BCC BY-NC 4.083.2K 57Oct 2024- Same category
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.
1.9BApache 2.01.1M 2.2KJan 2026- Same category
- Larger (1.9B vs 28M)
- Commercial use OK
Qwen3-TTS is a family of advanced multilingual, controllable, robust, and streaming text-to-speech models. Trained on over 5 million hours of speech data spanning 10 languages, Qwen3-TTS supports state-of-the-art 3-second voice cloning and description-based control.
915MApache 2.0839.5K 374Jan 2026- Same category
- Larger (915M vs 28M)
- Commercial use OK
Qwen3-TTS is a series of advanced multilingual, controllable, robust, and streaming text-to-speech models developed by the Qwen team.
906MApache 2.0715.7K 212Jan 2026- Same category
- Larger (906M vs 28M)
- Commercial use OK
ZONOS2 is our latest text-to-speech model trained on more than 6 million hours of varied multilingual speech, delivering expressiveness and quality on par with—or even surpassing—top TTS providers at low latency with MoE. ZONOS2 excels at high-fidelity and naturalistic voice cloning.
?BApache 2.0529 162Jun 2026- Same category
- Commercial use OK
FreyaTTS-small is a 183M-parameter Turkish text-to-speech model, and the open-source member of the FreyaTTS family. It is tokenizer-free at the character level (92 symbols, no phonemizer or G2P) and generates speech with a non-autoregressive conditional flow-matching DiT in the frozen AudioVAE2…
183MApache 2.04.5K 59Jul 2026- Same category
- Larger (183M vs 28M)
- Commercial use OK
We release Qwen3-TTS, a series of powerful speech generation models developed by Qwen, offering comprehensive support for voice cloning, voice design, ultra-high-quality human-like speech generation, and natural language-based voice control.
1.9BApache 2.0622.4K 485Jan 2026- Same category
- Larger (1.9B vs 28M)
- Commercial use OK
New: public adaptation toolkit Prepare data, audit train/validation splits, adapt a fixed voice or language, resume training, evaluate checkpoints, and export PyTorch or ONNX packages with the Inflect adaptation toolkit.
?BApache 2.0373 454Jun 2026- Same category
- Commercial use OK
Audio8 TTS 0.1B supports speech generation and zero-shot voice cloning. This repository contains the complete v4 mixed checkpoint, its neural audio codec, tokenizer, processor, and Hugging Face remote code.
170Maudio8-community-license-v1.01.3K 248Aug 2026- Same category
- Larger (170M vs 28M)
Built on LTX-2 by Lightricks. Dramabox is Resemble AI's expressive TTS, trained on top of the LTX-2.3 audio branch under the LTX-2 Community License. Huge thanks to the Lightricks team for open-sourcing the base.
?Bltx-2-community472 328Apr 2026- Same category
MeloTTS is a high-quality multi-lingual text-to-speech library by MIT and MyShell.ai. Supported languages include:
?BMIT140K 316Feb 2024- Same category
- Commercial use OK
New: public adaptation toolkit Prepare data, audit train/validation splits, adapt a fixed voice or language, resume training, evaluate checkpoints, and export PyTorch or ONNX packages with the Inflect adaptation toolkit.
?BApache 2.0257 141Jun 2026- Same category
- Commercial use OK
This model was introduced in SpeechT5: Unified-Modal Encoder-Decoder Pre-Training for Spoken Language Processing by Junyi Ao, Rui Wang, Long Zhou, Chengyi Wang, Shuo Ren, Yu Wu, Shujie Liu, Tom Ko, Qing Li, Yu Zhang, Zhihua Wei, Yao Qian, Jinyu Li, Furu Wei.
?BMIT279.7K 853Feb 2023- Same category
- Commercial use OK
Chatterbox Multilingual V3 is the latest general-purpose multilingual TTS model in the Chatterbox family. It keeps the same 0.5B model size while improving speaker similarity, reducing hallucinations, and producing more natural, conversational speech across languages.
500MMIT2.6M 1.9KApr 2025- Same category
- Larger (500M vs 28M)
- Commercial use OK
- Widely used
Fish Audio Drama 3 is Fish Audio's new-generation text-to-speech model, built for multi-character scenes and long-form content. It supports multi-speaker generation, voice cloning and natural-language control.
35.9BA4.7Bfish-audio-research-license44.6K 1.3KOct 2026- Same category
- Larger (35.9B vs 28M)