BWhat changed
Combined a Thinker-Talker mixture-of-experts system with multi-codebook speech prediction and a lightweight causal waveform decoder to start streaming from the first codec frame.
Scientific scope
Combined a Thinker-Talker mixture-of-experts system with multi-codebook speech prediction and a lightweight causal waveform decoder to start streaming from the first codec frame. Thinker-Talker is inherited from2.5Omni; the codec-frame/decoder and joint-training delta is distinct. Reported234ms is a theoretical cold-start figure, not guaranteed end-to-end service latency. Captioner/thinking/model sizes remain one package.
Evidence
The linked original sources describe the stated contribution and its evidence limits. WRS has not independently rerun the experiments or certified a proof. Thinker-Talker is inherited from2.5Omni; the codec-frame/decoder and joint-training delta is distinct. Reported234ms is a theoretical cold-start figure, not guaranteed end-to-end service latency. Captioner/thinking/model sizes remain one package.
Contributors
- Source-listed authors and technical contributors · SCIENCE: Source-listed authors or explicitly credited technical contributors for the cited versions. Family entries retain source-specific attribution; a combined list is not one original-paper byline or equal individual work. Names alone do not establish resolved identities.
- Qwen3-Omni Technical Report
- Qwen Team: multi-codebook first-frame streaming in omni models — official method source
Provisional WRS assessment, edition 2026-10-16.1. Independent expert calibration is pending. This grade does not certify correctness or assign ownership.
Full scientific record · Browse contribution grades · Request a correction