{"version":1,"type":"story","url":"https://digestai.news/story/google-deepmind-launches-gemini-3-8-live-and-3-8-live-extended-thinking-models","json":"https://digestai.news/story/google-deepmind-launches-gemini-3-8-live-and-3-8-live-extended-thinking-models.json","markdown":"https://digestai.news/story/google-deepmind-launches-gemini-3-8-live-and-3-8-live-extended-thinking-models.md","slug":"google-deepmind-launches-gemini-3-8-live-and-3-8-live-extended-thinking-models","headline":"Google launches Gemini 3.8 Live and Extended Thinking voice models","summary":"Google announced on September 15, 2026 the release of two speech‑to‑speech models—Gemini 3.8 Live and Gemini 3.8 Live Extended Thinking—designed for real‑time voice agents. The models support near‑real‑time visual grounding, automatic detection of 97 languages, and background tool or API calls that keep the conversation flowing. Gemini 3.8 Live Extended Thinking topped Artificial Analysis’ Speech‑to‑Speech Quality Index with a score of 82.6 and led on τ‑Voice (68.6%), Sierra’s τ‑Voice‑banking (35.1%) and Big Bench Audio (97.7%). Google listed partners such as Agora, Fishjam, LiveKit, Pipecat, Vercel and Vision Agents, and noted that all generated audio carries a SynthID watermark.\n\nMarktechpost alone provided pricing details of $0.005 per minute of audio input and $0.018 per minute of audio output, and added LangChain to the partner list. Socialmediaexaminer was the only source to mention a dedicated Windows 10/11 desktop app accessed via Alt + Space and to highlight additional Gemini features called Nano Banana and Gemini Omni. The ServiceNow EVA‑Bench benchmark was reported only by unite.ai.","keyPoints":["Google released Gemini 3.8 Live and Gemini 3.8 Live Extended Thinking on Sep 15 2026 for real‑time speech‑to‑speech voice agents.","Extended Thinking topped Artificial Analysis’ Speech‑to‑Speech Quality Index with 82.6 and scored 68.6% τ‑Voice, 35.1% τ‑Voice‑banking, 97.7% Big Bench Audio.","Marktechpost reports pricing of $0.005/min input and $0.018/min output; Socialmediaexaminer alone notes a Windows desktop app and Nano Banana/Gemini Omni features."],"whyItMatters":"The models bring near‑real‑time, multilingual voice interaction with background tool execution, enabling more natural AI assistants for enterprises and developers and setting new performance benchmarks in the voice‑agent market.","category":{"slug":"models","name":"Generative AI & Models","url":"https://digestai.news/category/models"},"entities":{"companies":["Google","Agora","Fishjam","LiveKit","Vercel","Vision Agents"],"models":["Gemini 3.8 Live","Gemini 3.8 Live Extended Thinking"],"people":["Tom Ouyang","Malini Jaganathan"]},"firstPublishedAt":"2026-09-15T17:05:57Z","updatedAt":"2026-09-17T13:02:21Z","sourceCount":7,"hasPrimarySource":true,"sources":[{"outlet":"Google DeepMind","title":"Introducing Gemini 3.8 Live and 3.8 Live Extended Thinking","url":"https://deepmind.google/blog/introducing-gemini-3-8-live-and-3-8-live-extended-thinking","publishedAt":"2026-09-15T17:05:57Z","type":"primary","primary":true,"lead":true},{"outlet":"Ben's Bites","title":"New home for Cowork","url":"https://bensbites.com/p/new-home-for-cowork","publishedAt":"2026-09-17T13:02:21Z","type":"newsletter","primary":false,"lead":false},{"outlet":"Social Media Examiner","title":"Creating AI Image Libraries, Showing Up in AI Results, and Industry News","url":"https://socialmediaexaminer.com/creating-ai-image-libraries-showing-up-in-ai-results-and-industry-news","publishedAt":"2026-09-17T09:30:00Z","type":"press","primary":false,"lead":false},{"outlet":"Simon Willison","title":"Gemini Live audio","url":"https://simonwillison.net/2026/Sep/15/gemini-live","publishedAt":"2026-09-15T22:47:07Z","type":"newsletter","primary":false,"lead":false},{"outlet":"MarkTechPost","title":"Google Releases Gemini 3.8 Live and 3.8 Live Extended Thinking for Production Grade Voice Agents","url":"https://marktechpost.com/2026/09/15/google-releases-gemini-3-8-live-and-3-8-live-extended-thinking-for-production-grade-voice-agents","publishedAt":"2026-09-15T20:53:09Z","type":"press","primary":false,"lead":false},{"outlet":"The Decoder","title":"Google launches Gemini 3.8 Live to take on OpenAI's GPT-Live-1 at a fraction of the cost","url":"https://the-decoder.com/google-launches-gemini-3-8-live-to-take-on-openais-gpt-live-1-at-a-fraction-of-the-cost","publishedAt":"2026-09-15T18:23:20Z","type":"press","primary":false,"lead":false},{"outlet":"Unite.AI","title":"Google Launches Gemini 3.8 Live and Extended Thinking Voice Models","url":"https://unite.ai/google-launches-gemini-3-8-live-and-extended-thinking-voice-models","publishedAt":"2026-09-15T17:24:01Z","type":"press","primary":false,"lead":false}],"sourceNotes":null,"discussions":[{"site":"Hacker News","url":"https://news.ycombinator.com/item?id=49715947","points":54}],"thread":{"title":"Google's Gemini 3.8 Voice Model Saga","url":"https://digestai.news/thread/google-deepmind-launches-gemini-3-8-live-and-3-8-live-extended-thinking-models","storyCount":4},"cite":{"text":"Digest AI, \"Google launches Gemini 3.8 Live and Extended Thinking voice models\", 15 September 2026, https://digestai.news/story/google-deepmind-launches-gemini-3-8-live-and-3-8-live-extended-thinking-models","publisher":"Digest AI","title":"Google launches Gemini 3.8 Live and Extended Thinking voice models","datePublished":"2026-09-15T17:05:57Z","url":"https://digestai.news/story/google-deepmind-launches-gemini-3-8-live-and-3-8-live-extended-thinking-models"},"generatedBy":"Written by Digest AI's editorial model from the linked sources; the sources are the record.","license":"Headlines, digests and key points are written by Digest AI and may be quoted with a link to the story page. Linked articles belong to their publishers. Terms: https://digestai.news/terms#reuse"}