Project AURIONAI Use-Cases Research & Innovation
speech synthesis

Higgs-Audio-v3-Studio

Higgs-Audio-v3-Studio hosts GGUF builds and a native C++/CUDA engine for a Windows desktop app that runs Higgs Audio v3 locally for TTS, voice cloning and multi-speaker generation; it is not the upstream release.

drbaph
Levelmodel
Classspeech synthesis
Parametersnot in the notes
Contextnot in the notes
Released2 Jul 2026
Opennessopen weights restricted
Runs onconsumer GPU, offline capable
Statusnot in the notes

Lineage

quantize
Boson higgs-tts-3-4b to Higgs-Audio-v3-Studio: quantize, medium confidenceBoson higgs-tts-3-4bBoson higgs-tts-3-4bHiggs-Audio-v3-StudioHiggs-Audio-v3-Studiodrbaph, Jul 2026
Boson higgs-tts-3-4b is the parent (quantize)medium confidencemodel card
Where this comes from
Parent and child relations from model cards, Hugging Face base_model fields and the genealogy tables of the research notes. The tree shows two generations each way, 2 models. Confidence and evidence are the notes' own.

R&D Topics

Large-scale zero-shot text-to-speechproduced by

Identity

Developer
drbaph
Family
not in the notes
Released
2 Jul 2026 sourcesregistry date
Weights published
2 Jul 2026
Record checked
11 Oct 2026
Where this comes from
Developer and dates from the research notes. Release date is the vendor's public announcement; weights date is when open weights were published.

Architecture and Size

Architecture
not in the notes
Total parameters
– sources
Active parameters
–
Checkpoint parameters
–
Context length
not in the notes
Input
text, audio
Output
audio
Where this comes from
Parameter counts as the developer states them. Where the checkpoint holds a different total, both are shown (register A16).

Openness and Access

Openness
open weights restricted
Licence
Boson Higgs TTS 3 Research and Non-Commercial License Agreement sources
Access routes
open weights download
Where this comes from
The licence file governs over registry tags and aggregator labels.

Deployment

Runs on
consumer GPU, offline capable
Footprint
GGUF builds q4_k_m (8 GB VRAM) to bf16 (16 GB); Windows x64 with NVIDIA RTX 30xx, 40xx or 50xx and a CUDA 13 driver
Runtimes
not in the notes
Where this comes from
Deployment tiers and runtimes as named in the notes. Footprint is the notes' figure, not a measurement by AURION.

Indonesia

Indonesian support
listed by developer
Regional languages
Javanese
Jakarta hosting
unknown
Languages
Card lists many languages including Indonesian, Javanese, Malay, Tagalog and Vietnamese
Where this comes from
Indonesia fields are carried on every model; unknown is an allowed value.

Benchmark Results

The notes record no benchmark result for this model.

Industry Use Cases

No industry use case links this model.

Sources

  1. huggingface.co/drbaph/Higgs-Audio-v3-Studio https://huggingface.co/drbaph/Higgs-Audio-v3-Studio
  2. huggingface.co/api/models/drbaph/Higgs-Audio-v3-Studio https://huggingface.co/api/models/drbaph/Higgs-Audio-v3-Studio

To cite this page: Project AURION, AI Use-Cases Research & Innovation, "Higgs-Audio-v3-Studio", https://aurion.semai.id/models/drbaph-higgs-audio-v3-studio/, records of October 2026.