--- title: Gemini 3.8 Live thinks in the background while you talk,... description: Google has released Gemini 3.8 Live and 3.8 Live Extended Thinking, models that can reason and run tool calls in the background without pausing a live... canonical: https://daily.dev/posts/gemini-3-8-live-thinks-in-the-background-while-you-talk-and-the-benchmarks-are-hard-to-ignore-vqety7y4w twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: Gemini 3.8 Live thinks in the background while you talk, and the benchmarks are hard to ignore | daily.dev og:description: Google has released Gemini 3.8 Live and 3.8 Live Extended Thinking, models that can reason and run tool calls in the background without pausing a live... og:url: https://daily.dev/posts/gemini-3-8-live-thinks-in-the-background-while-you-talk-and-the-benchmarks-are-hard-to-ignore-vqety7y4w og:image: https://api.daily.dev/og/posts/vqeTY7Y4w.png og:image:alt: Gemini 3.8 Live thinks in the background while you talk, and the benchmarks are hard to ignore og:image:width: 1200 og:image:height: 630 og:locale: en --- > ## Documentation Index > Fetch the complete documentation index at: https://daily.dev/llms.txt > Use this file to discover all available pages before exploring further. # Gemini 3.8 Live thinks in the background while you talk, and the benchmarks are hard to ignore **[Trends](https://daily.dev/sources/trends)** · 2 min read · 2 upvotes · 0 comments ## Summary Google has released Gemini 3.8 Live and 3.8 Live Extended Thinking, models that can reason and run tool calls in the background without pausing a live conversation. The architecture moves away from cascaded ASR-LLM-TTS pipelines toward a native end-to-end multimodal model, preserving tone and emotional cues that get lost at pipeline handoffs. Gemini 3.8 scores 82.6 on the Artificial Analysis Quality Index (top spot) and 35.1 on τ-banking agentic task completion. Pricing is $0.005/min input and $0.018/min output. New Live API features include async function calling, proactive audio, and send_client_content for injecting context mid-session. The model is available in Google AI Studio, the Gemini API, Search, and the Gemini app, with partner integrations in LiveKit, Pipecat, LangChain, and Vercel. Google also highlighted a 70+ language real-time translation demo in Google Meet, noting most Gemini users are non-English speakers. ## Content Google just dropped Gemini 3.8 Live and its Extended Thinking sibling, and the benchmark numbers are doing real work here. The model scores 82.6 on the Artificial Analysis Quality Index (first place), hits 68.6% on τ-Voice versus OpenAI's GPT-Live-1 Astra medium at 67.9%, and does all of this at $0.84/hour — roughly 80% cheaper than its nearest frontier competitor. The τ-Voice benchmark is worth understanding because it's not testing whether the model sounds pleasant. It runs the model through multi-step customer service scenarios across airline, retail, and telecom domains: hold a conversation, follow domain policies, call tools correctly, reach the right outcome. That's a harder bar than "does it sound natural," and Gemini 3.8 Live is clearing it at a price point that makes the cost comparison genuinely uncomfortable for OpenAI. The Extended Thinking variant adds something architecturally interesting: background reasoning that runs asynchronously while the conversation keeps moving. The model can kick off a multi-step tool call, narrate what it's doing, and stay responsive — no awkward silence while it thinks. Google demoed this with an SVG drawing task to show the quality gap between the standard model and the reasoning-enabled one, and the difference was visible. Other capabilities in the release: async function calling for long-running tools, proactive audio (the model only speaks when addressed or when something relevant happens), automatic language switching across 97 languages, and visual grounding so you can share your screen mid-conversation and the model uses it as context. Google Meet's real-time translation now covers 70+ languages on the same underlying stack. For developers, both models are live on Vercel's AI Gateway and Google AI Studio, with integrations in LiveKit, Pipecat, and LangChain. The API uses WebSockets with a realtime client — standard enough that dropping it into an existing voice agent project shouldn't be painful. The cost gap is the part that's going to drive adoption decisions. A narrow quality lead is easy to argue about. 80% cheaper is harder to dismiss. ## Questions this post answers ### What is new in Gemini 3.8 Live compared to previous voice models? Gemini 3.8 Live can reason and run tool calls in the background without pausing an ongoing conversation, using async function calling so tools execute while the user keeps talking. It also introduces proactive audio, which only responds when addressed or when something relevant occurs, and send_client_content, which injects context into a session without forcing a conversational turn. It scores 82.6 on the Artificial Analysis Quality Index and 35.1 on τ-banking agentic task completion. _Teams building voice agents can track new Gemini Live capabilities like these on daily.dev._ ### How much does Gemini 3.8 Live cost to use via the API? Gemini 3.8 Live is priced at $0.005 per minute of input audio and $0.018 per minute of output audio, positioned as competitive for real-time voice applications. It is available through Google AI Studio, the Gemini API, Google Search, and the Gemini app, with partner integrations in LiveKit, Pipecat, LangChain, and Vercel. _Compare real-time voice API pricing across providers before committing on daily.dev._ ### Why are native multimodal voice models better than cascaded ASR-LLM-TTS pipelines? Cascaded pipelines transcribe audio to text, reason over the text, then synthesize speech, losing paralinguistic information like tone, hesitation, and emotion at each handoff. A natively multimodal end-to-end model like Gemini 3.8 Live processes audio directly, preserving that information throughout the interaction rather than discarding it during transcription. _Developers evaluating voice AI architectures can follow shifts like this one on daily.dev._ ## Community take How the wider developer community reacted, aggregated from 2 discussions and 8 comments across x (as of 2026-09-15). **TL;DR:** Reactions are cautiously curious about background thinking and async tool calls, with people probing benchmark specifics, latency/turn-taking mechanics, and pricing, alongside some skepticism about reliability and availability gaps. **Sentiment:** 25% positive · 55% mixed · 20% skeptical **The case for** - The voice pricing is seen as genuinely undercutting call-center costs. - Combining live response with background thinking is viewed as solving the awkward silence problem in voice agents. - Topping the τ-banking-style benchmark is seen as tracking with real latency improvements. **The pushback** - Concern that a multi-second thinking pause in a voice call could feel like a dropped call. - Worry about accountability when a confident-sounding agent makes a wrong decision (e.g. booking the wrong flight). - Requests to fix AI slop and hallucinations suggest lingering distrust in output quality. - Lack of EU region support and older models still on paid plans are seen as gaps. **By community** - x (mixed): Replies mix genuine technical curiosity (benchmark scope, turn-taking, latency) with light skepticism about voice pauses, regional availability, and reliability. **Open questions** - How does the model perform specifically on the Speech to Speech Index rather than the text-focused Quality Index? - How should developers handle turn-taking when the model is thinking mid-sentence — pre-empt or wait for a pause? - When will EU region support and consumer app access (e.g. Gemini App with Plus) catch up? **Highlights** > @_philschmid Artificial Analysis describes the Quality Index as a primarily text and English language suite, with speech inputs benchmarked separately from it. Their native audio measure is the Speech to Speech Index, built on Big Bench Audio and tau-Voice. Where does 3.8 Live land there? > — [jatingargiitk on x](https://x.com/jatingargiitk/status/2099908846273863778) > @_philschmid An agent that keeps talking while it thinks in the background. So we finally shipped the confident coworker. Jokes aside, $0.005/min input is the first voice pricing that genuinely undercuts a call center. The question is who picks up when it confidently books the wrong flight. > — [daniel\_priscu on x](https://x.com/daniel_priscu/status/2099927014748610691) > @_philschmid live plus thinking is a fun combo. the pause while it thinks is the part i'd watch though, in voice a 3 second gap feels like the call dropped 🫠 > — [dusangran on x](https://x.com/dusangran/status/2099923678469238812) > @_philschmid Background thinking + async tool calls while still talking. That's the hard part of voice agents. τ-banking at #1 tracks, latency was the bottleneck. Curious how I'd wire turn-taking: pre-empt mid-sentence or wait for a pause? > — [miratechgeek on x](https://x.com/miratechgeek/status/2099934854498378068) > @_philschmid love this but pls get the guys at @googlecloud to support it in EU region too 😭😭 2.5 Live is pretty much on its way out already > — [hungquangphan on x](https://x.com/hungquangphan/status/2099918017681641703) **Source threads** - [x](https://x.com/_philschmid/status/2099908172899357093) · 0 points · 8 comments - [x](https://x.com/rohanpaul_ai/status/2099942230379401250) · 0 points · 0 comments --- Tags: [#ai-agents](https://daily.dev/tags/ai-agents), [#google-gemini](https://daily.dev/tags/google-gemini), [#multimodal](https://daily.dev/tags/multimodal), [#voice-ai](https://daily.dev/tags/voice-ai), [#google-deepmind](https://daily.dev/tags/google-deepmind) [View this post on daily.dev](https://daily.dev/posts/gemini-3-8-live-thinks-in-the-background-while-you-talk-and-the-benchmarks-are-hard-to-ignore-vqety7y4w) ```json {"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]} {"@context":"https://schema.org","@type":"TechArticle","headline":"Gemini 3.8 Live thinks in the background while you talk, and the benchmarks are hard to ignore","url":"https://daily.dev/posts/gemini-3-8-live-thinks-in-the-background-while-you-talk-and-the-benchmarks-are-hard-to-ignore-vqety7y4w","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/gemini-3-8-live-thinks-in-the-background-while-you-talk-and-the-benchmarks-are-hard-to-ignore-vqety7y4w"},"datePublished":"2026-09-15T17:25:07.729Z","dateModified":"2026-09-15T19:24:52.558Z","description":"Google has released Gemini 3.8 Live and 3.8 Live Extended Thinking, models that can reason and run tool calls in the background without pausing a live...","image":"https://i.ytimg.com/vi/3CyW24Pkz4o/sddefault.jpg","thumbnailUrl":"https://i.ytimg.com/vi/3CyW24Pkz4o/sddefault.jpg","isAccessibleForFree":true,"articleSection":"Trends","inLanguage":"en","publisher":{"@type":"Organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180}},"author":{"@type":"Organization","name":"Trends","logo":"https://media.daily.dev/image/upload/s--ZfSp3asX--/f_auto,q_auto/v1780996004/logos/trends?_a=BAMAMiWQ0","url":"https://daily.dev/sources/trends"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/gemini-3-8-live-thinks-in-the-background-while-you-talk-and-the-benchmarks-are-hard-to-ignore-vqety7y4w","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":2},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"ai-agents,google-gemini,multimodal,voice-ai,google-deepmind","timeRequired":"PT2M"} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"Trends","item":"https://daily.dev/sources/trends"},{"@type":"ListItem","position":3,"name":"Gemini 3.8 Live thinks in the background while you talk, and the benchmarks are hard to ignore"}]} {"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/gemini-3-8-live-thinks-in-the-background-while-you-talk-and-the-benchmarks-are-hard-to-ignore-vqety7y4w#faq","mainEntity":[{"@type":"Question","name":"What is new in Gemini 3.8 Live compared to previous voice models?","acceptedAnswer":{"@type":"Answer","text":"Gemini 3.8 Live can reason and run tool calls in the background without pausing an ongoing conversation, using async function calling so tools execute while the user keeps talking. It also introduces proactive audio, which only responds when addressed or when something relevant occurs, and send_client_content, which injects context into a session without forcing a conversational turn. It scores 82.6 on the Artificial Analysis Quality Index and 35.1 on τ-banking agentic task completion. Teams building voice agents can track new Gemini Live capabilities like these on daily.dev."}},{"@type":"Question","name":"How much does Gemini 3.8 Live cost to use via the API?","acceptedAnswer":{"@type":"Answer","text":"Gemini 3.8 Live is priced at $0.005 per minute of input audio and $0.018 per minute of output audio, positioned as competitive for real-time voice applications. It is available through Google AI Studio, the Gemini API, Google Search, and the Gemini app, with partner integrations in LiveKit, Pipecat, LangChain, and Vercel. Compare real-time voice API pricing across providers before committing on daily.dev."}},{"@type":"Question","name":"Why are native multimodal voice models better than cascaded ASR-LLM-TTS pipelines?","acceptedAnswer":{"@type":"Answer","text":"Cascaded pipelines transcribe audio to text, reason over the text, then synthesize speech, losing paralinguistic information like tone, hesitation, and emotion at each handoff. A natively multimodal end-to-end model like Gemini 3.8 Live processes audio directly, preserving that information throughout the interaction rather than discarding it during transcription. Developers evaluating voice AI architectures can follow shifts like this one on daily.dev."}}]} ```