Unverified: Google's Gemini 3.8 Live brings multi-step reasoning to realtime voice
xiaohu · x · 2026-09-16
A Chinese blogger claims Google released Gemini 3.8 Live and an Extended Thinking variant pushing realtime voice into multi-step reasoning (unverified; the version naming is questionable). Base version targets low-latency, high-concurrency deployments with automatic language switching across 97 languages; Extended Thinking adds synced "thinking while speaking," asynchronous background tool calls without interrupting conversation, and near-realtime visual understanding — e.g., playing chess from the board or turning a sketch into React code over voice chat.
More from Models
- ICML paper: finite context windows impose impossibility results for adversarial regret — burny_tech · 2026-11-02
- SenseTime open-sources SenseNova-Loop-9B: weight-shared 56-layer loop, 256K context, Apache 2.0 — AdinaYakup · 2026-11-02
- User says Gemini excels at long-form fanfic, producing stories up to 29 minutes of reading — Ok-Jello3739 · 2026-11-02
- Sam Altman calls tokens 'a kind of fake unit' that doesn't measure business value — rohanpaul_ai · 2026-11-02
- 'You Can Feel the Gradient Slopes': Dev Contrasts Benchmaxxed RL Models vs General Ones — theteknosaur · 2026-11-02
- LightOnOCR-3 Launches: Tops FRBench, Adds Layout Structure and Chart Extraction — IgorCarron · 2026-11-02