Higgs Audio

weights

open(Higgs TTS 3 (higgs-tts-3-4b) as safetensors on the Hub, ungated)

data

closed(no training-data statement for Higgs TTS 3)

code

partial(inference through third-party serving recipes

license

Boson-Higgs-TTS-3-Research-and-Non-Commercial-License(higgs-tts-3-4b, the current release of the generation line

separate-line

Higgs Audio v3 STT, Apache-2.0(four speech-to-text checkpoints (higgs-audio-v3-stt, -8b-stt and their v2 builds) released about ten weeks before Higgs TTS 3 in a collection of their own

superseded-release

higgs-tts-2-3b-base(Higgs Audio v2 (2025) and its tokenizer, still published under the Boson Higgs Audio 2 Community License (Llama 3 based, expanded license above 100,000 annual active users)

Higgs TTS 3, Boson's current text-to-speech release and the line its own repository now points to, is free to download but may not be used commercially without a separate license; a free grant lets creators publish and monetize content made with it. Boson's Apache-2.0 speech-to-text models are a separate line and do not change that. No training data or training code is published for Higgs TTS 3.

• https://huggingface.co/api/models/bosonai/higgs-audio-v3-8b-stt-v2 recorded 2026-09-27

  Hub record: tag "license:apache-2.0" and "gated":false.

• https://huggingface.co/api/models/bosonai/higgs-tts-3-4b recorded 2026-09-27

  Hub record: license_name "boson-higgs-tts-3-research-and-non-commercial-license", "gated":false, and model.safetensors in the file list.

• https://huggingface.co/bosonai/higgs-tts-2-3b-base/raw/main/LICENSE recorded 2026-09-27

  License titled "BOSON HIGGS AUDIO 2 COMMUNITY LICENSE AGREEMENT": "greater than 100,000 annual active users in the preceding calendar year, you must request an expanded license from Boson AI".

• https://huggingface.co/bosonai/higgs-tts-3-4b/raw/main/LICENSE recorded 2026-09-27

  License titled "BOSON HIGGS TTS 3 RESEARCH AND NON-COMMERCIAL LICENSE AGREEMENT": "Any Commercial use of the Higgs Materials requires a separate written license from Boson."; "is not an open source license".

• https://huggingface.co/bosonai/higgs-tts-3-4b/raw/main/README.md recorded 2026-09-27

  Card: "Released for research and non-commercial use under the **Boson Higgs TTS 3 Research and Non-Commercial License**."; recipes are in the "vLLM-Omni Higgs TTS 3 recipe".

• https://raw.githubusercontent.com/boson-ai/higgs-audio/HEAD/README.md recorded 2026-09-28

  README: "Higgs Audio v3 is here — you no longer need this repo!"; "**Higgs Audio v3** is a standalone release and does **not** depend on the code here."; the weights it links are "bosonai/higgs-audio-v3-tts-4b", which the Hub redirects to bosonai/higgs-tts-3-4b.

• https://www.boson.ai/blog/higgs-audio-v3-tts recorded 2026-09-28

  "Higgs TTS 3: beyond reading, toward real speech for voice AI", The Boson AI Team, June 4, 2026: "Higgs TTS 3 is built for voice chat: it speaks, not just reads." The post does not mention the speech-to-text models.

Hugging Face downloads summed over the declared TTS, tokenizer and speech-to-text checkpoints; the older v2 model still draws the most.

• https://huggingface.co/api/models?author=bosonai&limit=100 recorded 2026-09-27

  Trailing-30-day downloads: higgs-tts-2-3b-base 164,978, higgs-tts-3-4b 101,389, higgs-audio-v2-tokenizer 64,116, higgs-audio-v3-8b-stt-v2 8,295, higgs-audio-v3-stt 1,148.

Listeners in blind tests place Higgs TTS 3 in the lower third of the TTS arena, a step below Kokoro. Boson's separate speech-to-text model does much better on the English recognition board, but it is a different line.

• https://artificialanalysis.ai/text-to-speech/leaderboard recorded 2026-09-28

  Leaderboard row "Higgs Audio V3 TTS" (creator Boson AI, released Jun 2026, open weights at bosonai/higgs-audio-v3-tts-4b), rank 62 with "elo":1036.76, of 91 ranked models; Kokoro 82M v1.0 is 48th at 1065.

• https://huggingface.co/datasets/hf-audio/open-asr-leaderboard-results/resolve/d2c5b384deccdb82834f41aeaffcc618c00efa2f/english_short_latest.csv recorded 2026-09-27

  Results CSV row "bosonai/higgs-audio-v3-stt,4.3925".