Hy-SimulMT-30B-A3B
Tencent Hunyuan’s open real-time interpreter for meetings and speeches; also inside the Hy Translate app.
Model idtencent/Hy-SimulMT-30B-A3B
PricePrice not published
Checked27 October 2026
Where it stands
1newer speech and audio model since its release
- NVIDIA Nemotron 3.5 Transcribe (30B-A4B)28 Oct 2026
- Weights
- Weights are public under a permissive licence (Apache-2.0).
How to call it
3 ways to use Hy-SimulMT-30B-A3B, with the model id each provider expects.
| Provider | Model id | Endpoint | Docs |
|---|---|---|---|
| Hugging Face | tencent/Hy-SimulMT-30B-A3B |
huggingface.co | |
| Hugging Face (Interleave variant) | tencent/Hy-SimulMT-30B-A3B-Interleave |
huggingface.co | |
| GitHub (SimulInfer / SimulTrain) | github.com |
Pricing
Price not published.
No price source linked. Checked 27 October 2026.
Specs
- Input
- Audio
- Output
- Text and audio
- Context window
- Not published
- Max output
- Not published
- Open weights
- Yes
- Licence
- Apache-2.0
- Released
- 22 October 2026
- Status
- Current
- Type
- Speech and audio
What stands out
What its maker marketed at launch, and what people found later, each with its source.
Open simultaneous Chinese–English interpretation
Translates speech while the speaker talks (text plus speech via an external TTS, with voice cloning); vendor-reported StreamLAAL about 3 s on long-form RealSI audio.
Notes
30B MoE with ~3B active; FP8 and GGUF builds published. Speech output needs a configured TTS service. Benchmarks are Tencent’s own.