Google's Gemini Live API streams multimodal sessions over WebSockets; its stable native voice-agent models are Gemini 3.8 Live and Gemini 3.8 Live Extended Thinking.
Best fitDevelopers building low-latency conversational agents that need native audio, multimodal input, and function calling in a persistent WebSocket session.
Team sizesstartup; mid-market; enterprise; developer
Call surfacesWebSocket; server-side applications; client applications using ephemeral tokens
Pricing modelStandard Live API usage is token-metered: $3 per 1M input audio tokens ($0.005/min equivalent) and $12 per 1M output audio tokens ($0.018/min equivalent), plus separate text-token usage. Accumulated session context is billed each turn, so costs can grow with conversation history.
Pricing signalStandard paid rates are $0.005/min equivalent for input audio and $0.018/min equivalent for output audio, billed separately plus text tokens; Google also lists a free API tier.
Free plan or trialGemini API free tier listed; paid Live usage is token-metered.
Gemini Live API pricing and budget
Standard paid rates are $0.005/min equivalent for input audio and $0.018/min equivalent for output audio, billed separately plus text tokens; Google also lists a free API tier.
Standard Live API usage is token-metered: $3 per 1M input audio tokens ($0.005/min equivalent) and $12 per 1M output audio tokens ($0.018/min equivalent), plus separate text-token usage. Accumulated session context is billed each turn, so costs can grow with conversation history.
Gemini API free tier listed; paid Live usage is token-metered.
Catalog pricing checked September 28, 2026. Confirm current terms in the provider sources.
stable native voice-agent models: gemini-3.8-live and gemini-3.8-live-extended-thinking
bidirectional native audio
text, image, and video input
function calling
asynchronous tool execution
session controls and interruption handling
Runtime and fit
Voice stack
Native audio agents: Gemini 3.8 Live (gemini-3.8-live) and Gemini 3.8 Live Extended Thinking (gemini-3.8-live-extended-thinking).; Separate TTS models: gemini-3.8-flash-tts and gemini-3.8-flash-lite-tts generate speech; neither is a Live agent.; Separate transcription endpoints: gemini-3.5-transcribe handles audio files; gemini-3.5-transcribe-live is a Live API speech-to-text mode, not a conversational agent.; Separate translation mode: gemini-3.5-live-translate-preview is a Live API interpreter for 70+ languages, not a conversational agent; it is preview.
Telephony
Not publicly documented
Call controls
Not publicly documented
Integrations
Not publicly documented
Knowledge and data
Not publicly documented
Languages
The Live API documentation lists Japanese (ja) and multilingual support; native audio output can switch languages naturally in conversation. A supported-language list does not guarantee equivalent call quality. Separately, Transcribe lists Japanese (ja-JP) among 85+ languages; Flash TTS lists 130 languages and Flash-Lite TTS 101 (both include Japanese); Live Translate preview supports 70+ languages.
What stands out
The stable Gemini 3.8 Live models combine native audio dialogue with configurable tool calls over persistent WebSocket sessions; TTS, transcription, and translation are separate supporting models.
Considerations
Google labels the Live API preview even though both Gemini 3.8 Live model IDs are stable. Native audio sessions use AUDIO response modality; enable output transcription for text. Audio-only sessions are limited to 15 minutes and audio-plus-video sessions to 2 minutes unless extended with session management. Client-to-server apps need ephemeral tokens. Accumulated audio/context is billed per turn.
Before you decide
Questions about Gemini Live API
How should I budget for Gemini Live API?
The published pricing signal is: Standard paid rates are $0.005/min equivalent for input audio and $0.018/min equivalent for output audio, billed separately plus text tokens; Google also lists a free API tier. The recorded pricing model is: Standard Live API usage is token-metered: $3 per 1M input audio tokens ($0.005/min equivalent) and $12 per 1M output audio tokens ($0.018/min equivalent), plus separate text-token usage. Accumulated session context is billed each turn, so costs can grow with conversation history. Ask the provider for a current quote covering the same workload, channels, subscriptions, usage and implementation costs.
Can Gemini Live API use my existing data?
The listed knowledge and data sources include Not publicly documented. Confirm permissions, freshness and failure behavior for your exact systems with the provider.
What should I test in a Gemini Live API pilot?
Run representative calls with the same instructions and escalation rules. Record voice quality, completed tasks, handoffs and total cost. Check these cautions: Google labels the Live API preview even though both Gemini 3.8 Live model IDs are stable. Native audio sessions use AUDIO response modality; enable output transcription for text. Audio-only sessions are limited to 15 minutes and audio-plus-video sessions to 2 minutes unless extended with session management. Client-to-server apps need ephemeral tokens. Accumulated audio/context is billed per turn.
Compare Gemini Live with similar candidates
These products share recorded categories or company-size fit. Compare the actual channels, capabilities, and pricing before treating them as substitutes.
Programmable Voice component that connects a TwiML ConversationRelay session to an application WebSocket for STT/TTS, language settings, DTMF, and call events.
Telephony voice-agent runtime
Scope and signals
Best fit: Developers building their own voice-agent logic while keeping telephony, media handling, and call control in Twilio.
Voice stack: Cascaded speech interface with developer-supplied LLM
Choose the voice-agent model before designing the session
Google documents Gemini 3.8 Live and Gemini 3.8 Live Extended Thinking for native audio conversations. Pick an exact model ID and supported features before testing. Speech generation, transcription and live translation are separate modes; their availability does not establish agent behavior.
Plan for long calls, reconnects and interrupted tools
Google documents a 15-minute audio-only session limit and a two-minute audio-plus-video limit, with session-management techniques for extending sessions. Client applications need ephemeral tokens. Test a reconnect during a real task, a caller correction while a tool runs, and duplicate-action prevention in your application.
Pricing checked October 9, 2026. The standard paid Live API audio rates are $3 per million input tokens and $12 per million output tokens; Google also shows $0.005 and $0.018 per-minute equivalents. Input and output are separate, with additional text usage. These figures are model rates rather than a hosted phone-agent price.
Google bills the active session context again on each turn, including accumulated audio. Longer histories can therefore increase cost per turn; enabled transcriptions also add text-token charges. Compare short and long sessions using recorded token usage, and add transport, telephony, hosting and operational costs.
Test recognition, speech quality, interruptions, language switching, tool use, and human transfer with your own calls in the required language. Availability can vary by model, plan, and phone route.
DocsBot documents multilingual voice workflows and Japanese support. Verify the exact voice, phone path, and transfer behavior in your pilot. Review DocsBot language information ↗
Other providers’ language claims reflect cited vendor material; confirm Japanese and other required languages directly with each provider.
Listing does not imply endorsement or a performance guarantee. This directory is published by DocsBot. Editorial policy
Your business. Your shortlist.
Is Gemini Live API right for you?
Start with your website. Answer a few focused questions, then get a personalized PDF with fit, trade-offs, alternatives and a pilot plan for your call flow.
