# Adventists AI open-sources Sophon-Realtime v1 full-duplex voice model

> Original title: Adventists AI just open-sourced Sophon-Realtime v1 on ModelScope: a full-duplex voice model that hears how you speak, and answers with to...

- Source: ModelScope
- Published: 2026-11-09T13:00:00.000Z
- Bad Influence: https://badinfluence.news/items/ewlhynleckismao3zqwuamxy0
- Original: https://x.com/ModelScope2022/status/2119777024403444161

## Summary

Adventists AI open-sources Sophon-Realtime v1 on ModelScope, a full-duplex voice model that keeps listening while the user talks and replies about 0.4 seconds after they stop. The model can laugh, sigh, and clear its throat, and it scores 82.8 on VoiceBench and 81.8% on Big Bench Audio, ranking first among open full-duplex models. Its weights are released under Apache-2.0, and it reaches 48% Pass@1 on Full-Duplex-Bench v3 tool calling, ahead of Grok at 43%.

## Post

Adventists AI just open-sourced Sophon-Realtime v1 on ModelScope: a full-duplex voice model that hears how you speak, and answers with tone of its own.

• It keeps listening while you talk, answers ~0.4 s after you stop, and stops the moment you cut in
• It laughs, sighs and clears its throat when the conversation calls for it
• #1 among open full-duplex models on VoiceBench (82.8) and Big Bench Audio (81.8%)
• 48% Pass@1 on Full-Duplex-Bench v3 tool calling, ahead of Grok (43%)
Apache-2.0 weights. Built by @BKJinAI  and team.
https://modelscope.ai/collections/AdventistsAI/Sophon-Realtime-v1
https://www.modelscope.ai/studios/AdventistsAI/Sophon-Realtime

[Video](https://badinfluence.news/api/img-proxy?u=https%3A%2F%2Fpbs.twimg.com%2Famplify_video_thumb%2F2119710021244076032%2Fimg%2Fz78oBgP53LBVXBqS.jpg&mode=thumb&exp=1794528000&sig=753a70bda4566fac)