Text-to-speech
Higgs TTS 3 4B requirements, license & speed
By Boson AI · released 2026-06-04 · 4.65B parameters
Commercial useConditional
Min memoryNot stated
Runs on CPUUnknown
StreamingYes
Key facts
- License
- Boson Higgs TTS 3 Research and Non-Commercial License source
- Commercial use
- Conditional sourceNon-commercial, plus a Creator Use Grant for monetized content; products/APIs need a commercial license
- Parameters
- 4.65B sourcesafetensors parameter count
- Minimum memory
- Not stated by source
- Runs on CPU
- Unknown
- Streaming
- Yes sourceSSE streaming via SGLang-Omni / vLLM-Omni
- Voice cloning
- Yes source
- Languages
- 100+ languages source
- Released
- 2026-06-04 sourceHugging Face repo creation date (proxy for release date)
- Gated on HF
- No
Measured speed & memory
Benchmark pending. We haven't measured Higgs TTS 3 4B yet. Planned targets: Raspberry Pi 5, CPU-only x86, Apple Silicon Mac and consumer GPUs. Get notified.
Accuracy
We don't run accuracy evals. For quality and intelligibility, check the Hugging Face Open TTS Leaderboard.
Similar TTS models
Spotted an error? Report it. Every fact links to the model card or repo it came from.