Tencent Hunyuan’s Chinese–English real-time simultaneous translation model processes incoming audio continuously and streams translated text; an Interleave weight variant is also available.
Tencent Hunyuan officially released it on 2026-10-26 with a Chinese–English real-time translation demo; Apache-2.0 weights are available from the official Hugging Face organization. The checkpoint outputs translated text; speech playback and voice cloning require separately configured TTS and are not described as native model audio output.
Only published specifications are shown.
This record covers multiple tiers. Modalities and capabilities describe the group coverage; support and licensing depend on the individual tier.
— Tencent’s official Hugging Face repositories first published Hy-SimulMT-30B-A3B and Interleave variant weights under Apache-2.0. ↗ — Tencent Hunyuan formally released Hy-SimulMT for bidirectional Chinese–English simultaneous translation with streaming audio input and translated text output; speech playback and voice cloning require a separately configured TTS service. ↗
Hy-SimulMT extends Tencent Hunyuan’s model line into low-latency simultaneous translation, with Apache-2.0 weights first published on 2026-10-24 and an official launch on October 26; no weight inheritance from other Hunyuan models is inferred.
Tencent’s official Hugging Face repositories first published Hy-SimulMT-30B-A3B and Interleave variant weights under Apache-2.0.
View sourceTencent Hunyuan formally released Hy-SimulMT for bidirectional Chinese–English simultaneous translation with streaming audio input and translated text output; speech playback and voice cloning require a separately configured TTS service.
View sourceSelect a card to view the related model.
No related models recorded yet
Copy this template, add the field, proposed value, and primary source, and share it with the archive maintainer.