Skip to content

xAI · Voice

Grok Voice 3.0

xAI 🇺🇸VoiceGrok family
From Llmgateway's listing:
xAI's speech-to-speech voice agent model.
Listings
3
Providers
2

Evidence and sources

Released 14 Oct 2026Maker HQ: US (according to xAI's own website)
Sources: LiteLLM data, LLMGateway data, llminfo researchPrimary data from LiteLLM from LLMGateway: name, description from llminfo research: release date

Input & output

from the model's type
Text
Audio
Text
Audio

Capabilities

Supported operations and provider-stated features

Function callingAudio input

Listings

We don't have any priced listings for this model yet.

3 listings without a price are hidden.

Benchmarks

Named effort levels run from lowest to highest. Select a result to see its source details.

Speech

Artificial Analysis6 related benchmarks · 7 results
Time to First Audio
Reported by Artificial Analysis · via BenchBoard docs
Artificial Analysis live leaderboard · October 24 snapshot · Big Bench Audio prompts
Input Audio Cost on Big Bench Audio Subset
Reported by Artificial Analysis · via BenchBoard docs
Artificial Analysis live leaderboard · October 24 snapshot · 40-question Big Bench Audio subset
Speech Agent Arena · Elo
Reported by Artificial Analysis · via BenchBoard docs
Artificial Analysis live leaderboard · October 24 snapshot
Reported by SpaceXAI · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
October 13 release table · Speech Agent Arena snapshot
Speech Agent Arena · Task Success
Reported by SpaceXAI, Artificial Analysis · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
October 13 release table · blind-call task success
Speech to Speech Quality Index
Reported by SpaceXAI, Artificial Analysis · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
October 13 release table · Artificial Analysis Speech to Speech Index
τ-Voice · Agentic Performance
Reported by SpaceXAI, Artificial Analysis · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
October 13 release article · Artificial Analysis τ-Voice customer-support tasks

Other

Data credits: BenchBoard by SYuan03 (github.com/SYuan03/benchboard). MIT covers its code and data organization; no results licence is inferred. Numeric relays retain the original stating-source links. · Data by LLM Stats

Around the web

Also known as: grok-voice-3-0, xai/grok-voice-3.0

Like this one. Tap or hover any dotted value, label or icon to see where it comes from and how it is measured. open details when tapped or hovered. How this data works