Gemini 3.8 Live
Among the best-liked voice agent models, free to start and the cheapest of the top group. Weak at hard support tasks.
Should you use it?
Pick it if
Developers building voice agents and phone bots who want top-tier quality at a low price.
Best for: Voice agents and phone bots that need top quality at a low price (Voice chat Top 20)
Skip it if
Your agent must finish long, multi-step support tasks; GPT-Live-1 completes 74.5% of a customer service test, against 30.1% here.
You give up: Finishes only about 30% of hard support tasks; free tier data improves Google products
The short version
- What this is
- Google's real-time voice model for apps that talk with people: it listens to speech and answers in speech, through the Gemini Live API. A thinking variant handles harder requests.
- Why it matters
- It is one of the best-rated voice agent models and costs a fraction of its closest rivals.
- 2nd of 26 on Artificial Analysis's voice agent arena; no score yet, as its terms stop us publishing one.
- About $0.023 per minute of conversation on paid API rates, with a free tier.
- Use the paid tier if you don't want inputs used to improve Google products.
Good at
Our take
Gemini 3.8 Live became generally available on 15 Sep 2026, alongside an Extended Thinking variant s4. In Artificial Analysis's blind arena, where people talk to two unnamed voice agents and pick the better one, it ranks 2nd of 26 s1. It also costs far less than the other top models s2.
The trade-off is depth. The standard model resolves only about 30% of the τ-Voice customer service scenarios; the Extended Thinking variant more than doubles that but is less liked in the arena s1 s2. Pick the standard model for fast, natural conversation, and test the thinking variant if your agent has to work through multi-step tasks.
Score
A score from 0 to 100 per job, against everything available today.
New to this? What the scores and labels mean
- 0 to 100
- Our score for one job, like writing or coding, compared with everything else out today. 100 is the best available today. For AI models we build it from public data, such as Arena's blind votes and Epoch AI's independent tests; apps and tools are scored on our own checklist. The same tool can score differently for different jobs, and where two likely ranges overlap, we call it a tie.
- Our verdict
- Use it: a top choice for most people. Worth trying: good, and worth a go. Depends: good for some people or one kind of job. Wait: too early to tell, or not out for everyone yet. Skip: something else does the job better.
- First look
- Just released. We say what it is and whether to bother, and give it a score once public results exist.
- No score yet
- None of the public sources we use covers it yet, or the only evidence is from a source whose numbers we can't use, so we describe it in words instead.
- Best buy
- Nothing cheaper scores as well for that job.
- Best free
- The best score for that job that costs nothing, through a free plan or by running on your own computer.
- Pay as you go
- No monthly plan: you pay for each use, usually through a developer account.
- Checked
- The day we last confirmed the score and the price against the sources. Prices change often, so check the date.
- AI model
- The engine underneath, like GPT or Claude. You use it through an app or, if you code, directly.
- Open source
- Free to download and run on your own computer, so nothing leaves your machine.
Voice and audio
No score yet. The leaderboard this rests on does not let us publish numbers from it, so we describe the evidence instead.
2 sources
Public leaderboardIndependent testArtificial Analysis Speech Agent Arena rank 2 of 26, range 1 to 4, Elo 1083, 482 samples, not marked preliminary s1; the 26 entries are about 21 distinct models, so the small-board rule does not apply. Independent check: Artificial Analysis Speech to Speech Index rank 2 of distinct models (Gemini 3.8 Live Extended Thinking High, 82.6; GPT-Live-1 is 1st).
Alternatives
Nothing we have scored beats it for voice on score or on price.
- Eleven v4 TurboVoice model–FreeIf you need text to speech rather than a full voice conversation model. Side by side
- Gemini 3.8 Flash TTSVoice model–FreeGoogle's text to speech model, for narration and read-aloud. Side by side
- GPT-Live 1Voice model–Pay as you goFirst on the Speech to Speech Index and stronger on support tasks, at $4.47 to $5.83 per audio hour. Side by side
Evidence
Everything behind the call and the scores, folded so the page stays short. Spotted a wrong price, score or link? Report a mistake.
How we scored this
Voice has no score yet, because none of the public sources we use covers it.
Public leaderboardIndependent test
Reviews are drafted with help from AI models, including Anthropic's Claude. Model scores come only from outside leaderboards and independent evaluations, by a written rule.
No one paid for this review or gave us free access, and we earn nothing if you sign up. How we score
Sources 4
- s1Artificial Analysis Speech Agent Arena leaderboard
- s2Artificial Analysis speech to speech leaderboard
- s3Gemini API pricing
- s4Gemini API changelog
Something here out of date or wrong? Report a mistake and we will check it against the source.
History 2 changes
Every change to a rating, the price or our call, with the reason. Scores are against what exists on the day, so they drift down as better things ship. Older entries come from before we moved to scores.
- New: Voice ratedReviewed: voice.
- Audit: re-opened the Speech Agent Arena and Speech to Speech pages and confirmed the figures. Added GPT-Live-1, the closest rival, as an alternative in place of Scribe v2; evidence notes the distinct-model count. Rating unchanged.
Use it well 4 guides on this site
Pages on this site that help with the jobs above.