- Start
- Sep 15, 202690% CONFIDENCEfrom the source
Gemini 3.8 Live and 3.8 Live Extended Thinking Released
- Google introduced Gemini 3.8 Live and Gemini 3.8 Live Extended Thinking on 15 September 2026 as "our most advanced live dialogue models yet": 3.8 Live is "built for scale and cost efficiency" and Extended Thinking "for high-complexity tasks, with increased intelligence and multi-step reasoning"[1]
- Extended Thinking "reasons and speaks simultaneously", using early cues like "Let me check that…" and narrating progress while multi-step tasks run in the background[1]
- 3.8 Live reads visual input in near real time, detects and switches among 97 supported languages mid-conversation, and keeps talking while tool and API calls finish in the background[1]
- Both reached developers in the Gemini API and Google AI Studio and enterprises in private preview in Gemini Enterprise; 3.8 Live rolled out in Search Live and Extended Thinking in Gemini Live, Docs Live, Gmail Live and Keep Live[1][5]
- They follow Gemini 3.1 Flash Live, released in March[5]
- On 24 September Google added Live Avatar, a near real-time video persona for 3.8 Live, in Gemini Enterprise[4]
Notable features
- Extended Thinking takes the top spot on Artificial Analysis' Speech to Speech Quality Index (82.6) and scores 68.6% on τ-Voice, 35.1% on Sierra's τ-Voice-banking benchmark and 97.7% on Big Bench Audio; 3.8 Live places second in the Speech Agent Arena[1]
- Prices per million tokens: $0.75 for text, $3.00 for audio and $1.00 for image or video input, and $4.50 for text and $12.00 for audio output - about $0.005 a minute of audio in and $0.018 out[3]
- Takes audio, images, video and text, with a 131,072-token input limit and 65,536 output tokens, and supports asynchronous function calling and Google Search grounding[2]
- Based on Gemini 3 Pro, with a January 2025 knowledge cutoff[6]
- All generated audio carries an imperceptible SynthID watermark[1]
- Voice platforms including Agora, LangChain, LiveKit, Pipecat and Vercel support it through the Gemini Live API, and Salesforce, Genspark and Lumeris are launch partners[1]
References 687% CONFIDENCEOverall confidence: 87%How well the pin's source and references back up its dates.Weighted average of how firmly 6 references, the source included, support the pin's start and end times; a reference counts half as much for every 180 days older than the newestShow all pins at 75% confidence or better
The first entry is always the pin's source. Overall confidence is a weighted average of how firmly each reference supports the start and end times used above; a reference counts half as much for every 180 days older than the newest.
- [1]90%blog.google/innovation-and-ai/models-and-research/gemini-models/gemini-3-8-live-gemini-3-8-live-extended-thinkingblog.google· Posted Sep 28, 2026· Starts Sep 15, 2026 ✓· 17% of score
Google's[2][3] post is dated "Sep 15, 2026" and says "3.8 Live is rolling out starting today"; 9to5Google's[5] report of the same day says "Google today announced Gemini 3.8 Live and Gemini 3.8 Live Extended Thinking", and the Gemini API changelog lists both as generally available on September 15.
- [2]90%Gemini 3.8 Live Extended Thinking | Gemini APIai.google.dev· Added Sep 28, 2026· 17% of score
The API model page for gemini-3.8-live-extended-thinking: text, image, audio and video in, text and audio out, 131,072 input and 65,536 output tokens, asynchronous function calling and search grounding supported, latest update September 2026.[1]
- [3]85%Gemini Developer API pricingai.google.dev· Added Sep 28, 2026· 17% of score
The pricing page lists Gemini 3.8 Live and 3.8 Live Extended Thinking together: $0.75 text, $3.00 audio and $1.00 image/video input and $4.50 text and $12.00 audio output per million tokens, or about $0.005 and $0.018 a minute of audio.[1]
- [4]75%Introducing Gemini 3.8 Live with Live Avatarblog.google· Published Sep 24, 2026· 17% of score
Google's[2][3] follow-up post of 24 September adds Live Avatar to 3.8 Live, pairing the voice model with near real-time video of a speaking persona, available in Gemini Enterprise.
- [5]90%Gemini 3.8 Live Extended Thinking powers Gemini Live, Gmail, & Keep9to5google.com· Published Sep 15, 2026· Starts Sep 15, 2026· 16% of score
9to5Google's same-day report by Abner Li that Google[2][3] "today announced" both models, following 3.1 Flash Live in March; it repeats the benchmark claims and says Extended Thinking powers Gmail Live, Docs Live and Keep Live and is rolling out to Gemini Live, while 3.8 Live powers Search Live.
Suggest a correction
Something missing or wrong? Say it in your own words: a link that backs this pin up, a different start or end date and why, or a fact it lacks or gets wrong. The AI checks it against this pin's sources, searches for better ones, and adds any page that backs you up. The pin's own sources still count most. A picture that shows something else, or shows it badly, is looked at too, and moved down or replaced.