whatmodel.sh / tts / Higgs Audio V3 TTS
Higgs Audio V3 TTS
↗ share
+ contribute
Higgs Audio V3 TTS is an open-weights text-to-speech model from Boson AI, released June 2026. Its published size is 4B parameters. Licence: boson-higgs-audio-v3-research-nc.
Score within the Vapi Humanness · Bradley–Terry rating, 2026-09-25 snapshot comparison set only.
$ how it compares
The text-to-speech models ranked either side of Higgs Audio V3 TTS, out of 148 with a score, then more from Boson AI.
| # | model | score | where | ||
|---|---|---|---|---|---|
| 21 |  Eleven v3 | 65±1 | api $100.00/M chars | ||
| 22 |  Chirp 3: HD | 65±4 | api $30.00/M chars | ||
| 23 | Higgs Audio V3 TTS | 65±3 | api $15.00/M chars +local | ||
| 24 |  Raon-OpenTTS-0.3B | 65±2 | comparison only | ||
| 25 |  Qwen3-TTS 12Hz 1.7B Base | 65±1 | local | ||
| more from Boson AI | |||||
| 66 | Higgs Audio V2 | 55±3 | local | ||
see all text-to-speech models · find the ones that fit your machine
$ what it's good at
voice cloning▁·▁▁▁▁▁▁▇█▃▃ #9 of 37 * · 37% of its score · 1 test
Beats Gemini 3.1 Flash TTS Preview, OmniVoice here, though they rank higher overall. Trails VoxCPM2 here, though they rank lower overall.
| test | result | vs the others in its run | counts |
|---|---|---|---|
| Seed-TTS-Eval · test-en/test-zh macro mean error lab | 1.1% ✎ | ▄··▄▄····█▄▄ beats 6/6 | 37% |
other tests6 not in a task · 58% of its score
| test | result | vs the others in its run | counts |
|---|---|---|---|
| RW-Voice-EQ TTS · Language Code-Switching (human) | 3.38 / 5 ✎ | █··▄·······▄ beats 2/3 | 20% |
| RVCBench LibriTTS clean voice cloning · ECAPA speaker similarity | 56.0% ✎ | ▃·····▂▂▂▃█▆ beats 7/15 | 16% |
| MiniMax multilingual test set · 23-language mean error lab | 2.7% ✎ | ▃·▃·▃·····▃█ beats 6/6 | 8.1% |
| Arabic TTS Arena · community blind A/B rating, 2026-09-24 snapshot | 1072.8 ✎ | ▂▂··▃·█▅···▂ beats 11/12 | 7.7% |
| CV3-Eval multilingual voice clone · 9-language mean error lab | 4.4% ✎ | ▃·····▃···▅█ beats 6/6 | 5.9% |
| RVCBench LibriTTS clean voice cloning · Whisper-medium WER | 5.0% ✎ | ▂▂▂··▂·▃▃▃█▂ beats 11/15 | 0.2% |
$ where to run it
Through an API
api · $15.00/M chars · docs.boson.ai ↗ · price sourceUSD per 1M characters: Boson AI lists Higgs TTS 3 text input at $0.015 / 1K characters; generated audio free (x1000). Prepaid usage-based credits. Product: Boson API. API model: Higgs TTS 3. Settings: standard. Price basis: published list rate. Exclusions: taxes. Currency: USD. Boson AI official pricing docs, observed 2026-09-25.
Other ways to run it
version
- Run it as published · needs ~10.1 GB, NVIDIA only · from github.com/boson-ai (official)
hf download bosonai/higgs-audio-v3-tts-4b && sgl-omni serve --model-path bosonai/higgs-audio-v3-tts-4b --port 8000
Official weights and documentation: huggingface.co/bosonai ↗
$ how the score was worked out · Higgs Audio v3 comparison tab… 51%, RW-Voice-EQ Bench (Hume AI) 20%, RVCBench: Benchmarking Robust… 16% and 1 more
Each result implies a score on its own, after adjusting for how hard its test is and who ran it. The score is their weighted average, with a small pull toward the middle for thinly tested models. Each test's weight is its counts column above; here it's added up by who ran the tests. How it works
| source | results | share of the score |
|---|---|---|
 Higgs Audio v3 comparison tables in SGLang-Omni cookbook (Boson AI contribution) · lab-reported | 3 | 51% |
 RW-Voice-EQ Bench (Hume AI) | 1 | 20% |
| RVCBench: Benchmarking Robustness of Voice Cloning Across Modern Audio Generation Models (arXiv:2602.00443v2) | 2 | 16% |
| Arabic TTS Arena (Navid AI) · leaderboard snapshot 2026-09-24 | 1 | 8% |
| prior · pull toward the middle | 5% | |
| weighted average | = 65.0 |
$ improve this page
Use a short GitHub form or edit the data through a pull request. Every number needs a source; here's which ones we can use.
data updated · sources and licences · json