All models
Voice & realtimeLive
SpaceXAIGrok Voice 3.0
Grok's realtime voice model for spoken conversations, customer support and assistants that can use tools during a call.
Plain-English overview
What Grok Voice 3.0 actually is
It listens and speaks within one live session, rather than only reading a script aloud. Applications can connect business tools and search to help answer questions or complete tasks.
Reasoning is enabled by default and can be disabled. Pin grok-voice-3.0 for a specific version; grok-voice-latest follows future upgrades.
Good fit for
- Customer-support calls
- Multilingual voice assistants
- Voice agents connected to business tools
Category comparison
The facts that matter for voice models
These are provider-published specifications, not Cody benchmark scores. Follow the linked sources for current limits and endpoint-specific exceptions.
- Price basis
- $0.08 / audio minuteToken-, character-, or minute-based price for the listed access route.
- Interaction mode
- Realtime speech-to-speech with toolsSpeech-to-speech, audio-to-audio, or streamed text-to-speech behavior.
- Published latency
- Sub-second provider claimProvider-published measurement with its stated exclusions; not a Cody test.
- Languages
- 20+ languagesProvider-documented language coverage or a clear unpublished marker.
- Tools & control
- Functions, web/X/file search and remote MCPNative function calling, reasoning, or speech-control features.
- Context or input limit
- Not publishedDocumented token or character limit where it is meaningful.
API and provider access
Where to get Grok Voice 3.0
Availability
Regions and access stage
Available in the xAI Realtime API as grok-voice-3.0. The grok-voice-latest alias selects this version from October 14, 2026.
Account and regional availability should be verified in the SpaceXAI console.
Check live availabilityData and training
The route matters.
SpaceXAI says it does not train on API inputs or outputs without explicit permission. Default encrypted retention is 30 days; team-level zero data retention is available with feature tradeoffs.
This is a concise reading of the cited provider material, not legal advice. A third-party gateway can have different storage, routing, training, and residency terms from the model maker's direct API.
Read the provider policyFrequently asked questions
Grok Voice 3.0 FAQ
What is Grok Voice 3.0?
Grok's realtime voice model for spoken conversations, customer support and assistants that can use tools during a call. It listens and speaks within one live session, rather than only reading a script aloud. Applications can connect business tools and search to help answer questions or complete tasks.
When was Grok Voice 3.0 released?
Grok Voice 3.0 was released on October 14, 2026 according to the cited provider materials.
Where can I access Grok Voice 3.0?
Available in the xAI Realtime API as grok-voice-3.0. The grok-voice-latest alias selects this version from October 14, 2026. The access routes listed in this guide are SpaceXAI.
How much does Grok Voice 3.0 cost?
$0.08 / audio minute. USD 4.80 per audio hour, plus USD 0.004 per text input. Tool and telephone-service charges can add to the total; this is not a token-based rate.
Where is Grok Voice 3.0 available?
Available in the xAI Realtime API as grok-voice-3.0. The grok-voice-latest alias selects this version from October 14, 2026. Account and regional availability should be verified in the SpaceXAI console.
Is my Grok Voice 3.0 API data used for training?
SpaceXAI says it does not train on API inputs or outputs without explicit permission. Default encrypted retention is 30 days; team-level zero data retention is available with feature tradeoffs. The policy belongs to the provider route and account terms, so verify it again before production use.
Related comparisons
Head-to-head comparisons
Grok Voice 3.0 vs GPT-Live 1
OpenAI's full-duplex voice frontend for natural conversation that delegates deeper reasoning and actions to a separate backend agent.
Open full comparisonGrok Voice 3.0 vs Gemini 3.8 Live
Google's real-time voice model for responsive conversations with audio, text and visual context.
Open full comparisonGrok Voice 3.0 vs Grok Voice 2.0
Previous-generation Grok voice model, deprecated in favor of Grok Voice 3.0.
Open full comparison
