Top open text-to-speech models
Models that convert text into natural-sounding speech.
How this list works: ranked by Hugging Face downloads in the last 30 days, among models with 1,000+ downloads.
| # | Model | Author | Downloads (30 days) | Likes | Size | License | Commercial use | Runs on |
|---|---|---|---|---|---|---|---|---|
| 1 | Kokoro-82M | hexgrad | 10.5M | 7.3K | 82M | Apache 2.0 | Yes | CPU or GPU |
| 2 | audio.cpp-gguf | audio-cpp | 6.4M | 255 | – | Custom (other) | Check the license | CPU |
| 3 | XTTS-v2 | coqui | 6.3M | 3.9K | – | Custom (other) | Check the license | CPU or GPU |
| 4 | Qwen3-TTS-12Hz-1.7B-CustomVoice | Qwen | 2.1M | 2.1K | 1.7B | Apache 2.0 | Yes | GPU |
| 5 | OmniVoice | k2-fsa | 1.7M | 1.6K | – | Not stated | Check the license | GPU |
| 6 | chatterbox | ResembleAI | 1.7M | 1.9K | – | MIT | Yes | CPU or GPU |
| 7 | VieNeu-TTS-v3-Turbo | pnnbao-ump | 1.3M | 82 | – | Apache 2.0 | Yes | CPU |
| 8 | kokoro-inno-clone-tuner | remsky | 1.1M | 26 | – | Apache 2.0 | Yes | GPU |
| 9 | Qwen3-TTS-12Hz-0.6B-CustomVoice | Qwen | 993.7K | 206 | 600M | Apache 2.0 | Yes | GPU |
| 10 | F5-TTS | SWivid | 846.2K | 1.2K | – | CC BY-NC 4.0 | No | CPU or GPU |
| 11 | VoxCPM2 | openbmb | 819.8K | 1.7K | – | Apache 2.0 | Yes | GPU |
| 12 | Kokoro-82M-v1.0-ONNX | onnx-community | 695.4K | 276 | 82M | Apache 2.0 | Yes | CPU |
| 13 | Qwen3-TTS-12Hz-0.6B-Base | Qwen | 611.1K | 340 | 600M | Apache 2.0 | Yes | GPU |
| 14 | VibeVoice-1.5B | microsoft | 552.4K | 2.5K | 1.5B | MIT | Yes | GPU |
| 15 | sanoTTS | ampixa | 490.8K | 70 | – | GPL-3.0 | Yes, with conditions | CPU |
| 16 | Qwen3-TTS-GGUF | Serveurperso | 465.7K | 60 | – | Apache 2.0 | Yes | CPU |
| 17 | Qwen3-TTS-12Hz-1.7B-VoiceDesign | Qwen | 401.6K | 443 | 1.7B | Apache 2.0 | Yes | GPU |
| 18 | Higgs-Audio-v3-Studio | drbaph | 390.6K | 7 | – | Custom (other) | Check the license | CPU |
| 19 | VibeVoice-Realtime-0.5B | microsoft | 292.3K | 1.3K | 500M | MIT | Yes | GPU |
| 20 | Fun-CosyVoice3-0.5B-2512 | FunAudioLLM | 231.9K | 666 | 500M | Apache 2.0 | Yes | CPU |
| 21 | tts-hifigan-libritts-22050Hz | speechbrain | 217.7K | 6 | – | Apache 2.0 | Yes | CPU or GPU |
| 22 | indic-parler-tts | ai4bharat | 206.4K | 427 | – | Apache 2.0 | Yes | GPU |
| 23 | AuK-Base-and-Flash-GGUF | audio-cpp | 204.4K | 6 | – | MIT | Yes | CPU |
| 24 | Drama-3-GGUF | Jmica | 164.8K | 22 | – | Custom (other) | Check the license | CPU |
| 25 | MeloTTS-English | myshell-ai | 163K | 316 | – | MIT | Yes | GPU |
Other model comparisons
Top open language modelsTop open image modelsTop open audio modelsTop open video modelsTop open embedding modelsTop open multimodal modelsTop open text-generation models (LLMs)Top open text-to-image modelsTop open speech-recognition modelsTop open feature-extraction modelsTop open sentence-similarity modelsTop open image-to-text modelsTop open vision-language modelsTop open image-classification modelsTop open image-segmentation modelsTop open object-detection modelsTop open depth-estimation modelsTop open audio-classification modelsTop open video-classification models