As of: 2026-11-13 13:49 CET.
Researched and written by AI agents (Claude Opus 5.5 in Claude Code), with a human editor. How we work: https://postcutoff.com/about/
Canonical page: https://postcutoff.com/m/grok-voice-think-fast-2-0/

# Grok Voice Think Fast 2.0

xAI, Grok Voice family. Status: legacy. Type: speech and audio. Released: 2026-07-29.

## How to call it

| Provider | Model id | Endpoint | Docs |
|---|---|---|---|
| xAI API (Voice Agent / speech-to-speech, WebSocket) | `grok-voice-think-fast-2.0` | wss://api.x.ai/v1/realtime?model=grok-voice-think-fast-2.0 | https://docs.x.ai/developers/model-capabilities/audio/voice-agent |
| xAI API (alias) | `grok-voice-latest` |  | https://docs.x.ai/developers/model-capabilities/audio/voice-agent |
| Web app |  |  | https://grok.com |

## Pricing

- Per minute: $0.08
- Unit as published: USD per minute of audio ($4.80/hr), plus $0.004 per text input (as listed on the xAI pricing page)
- Source: https://docs.x.ai/developers/pricing
- Last checked: 2026-10-14

## Specs

- Input: audio, text
- Output: audio, text
- Context window: not published
- Max output: not published
- Open weights: no
- Licence: proprietary

## Where it stands

- Newer speech and audio models in the registry since its release: 39. The 5 newest:
  - NVIDIA Nemotron 3.5 Transcribe (30B-A4B) (2026-10-28) https://postcutoff.com/m/nemotron-3-5-transcribe/
  - Hy-SimulMT-30B-A3B (2026-10-22) https://postcutoff.com/m/hy-simulmt-30b-a3b/
  - Eleven Realtime v1 Max / Mini (2026-10-20) https://postcutoff.com/m/eleven-realtime-v1-max/
  - Voxtral Mini Transcribe 3 (+ Mini Transcribe Realtime 3) (2026-10-20) https://postcutoff.com/m/voxtral-mini-transcribe-3/
  - Voxtral Small Transcribe 3 (2026-10-20) https://postcutoff.com/m/voxtral-small-transcribe-3/
- Before it from xAI: Grok Text to Speech (Grok TTS API) (2026-04-17) https://postcutoff.com/m/grok-tts/
- After it from xAI: Grok Voice Transcribe 2.0 (2026-09-18) https://postcutoff.com/m/grok-voice-transcribe-2-0/
- Closed weights; access is hosted only.

## What stands out

- Reasoning while speaking: Speech-to-speech model that reasons in real time (reasoning effort 'high' by default, can be set to 'none'); 97.2% Big Bench Audio, 82.9 on the Artificial Analysis Speech-to-Speech Quality Index (vs 75.7 for v1.0). Source: https://x.ai/news/grok-voice-think-fast-2
- Faster first audio: Time to first audio cut from 1.25 s (v1.0) to 0.70 s; Full Duplex Bench 95.1%, tau-voice Bench 56.5% (xAI-reported). Source: https://x.ai/news/grok-voice-think-fast-2
- Built-in server-side tools: Web search, X search, collections (file) search and remote MCP callable from inside a voice session, plus custom functions. Source: https://docs.x.ai/developers/model-capabilities/audio/voice-agent
- OpenAI Realtime-compatible protocol: Largely compatible with the OpenAI Realtime SDK: change base URL to https://api.x.ai/v1 and the API key (minor event-name differences). Source: https://docs.x.ai/developers/model-capabilities/audio/voice-agent

## Notes

Superseded by Grok Voice 3.0 on 2026-10-14 (grok-voice-latest now points to 3.0; see entry 2026-10-14-grok-voice-3). Released 2026-07-29; grok-voice-latest switched to it on 2026-08-05. Predecessor grok-voice-think-fast-1.0 can still be pinned. 20+ languages; audio PCM (8-48 kHz), Opus 24 kHz, G.711 mu-law/A-law; server VAD, session resumption (30 min), custom cloned voices. xAI says Starlink A/B tests raised sales conversion and support containment. Benchmarks are xAI-reported.

xAI's realtime voice-agent model (also used in the Grok app, Tesla vehicles and Starlink support).

```python
# OpenAI Realtime-compatible WebSocket
import websockets, os
url = "wss://api.x.ai/v1/realtime?model=grok-voice-think-fast-2.0"
ws = await websockets.connect(url, additional_headers={"Authorization": f"Bearer {os.environ['XAI_API_KEY']}"})
```

Sources: https://x.ai/news/grok-voice-think-fast-2 · https://docs.x.ai/developers/model-capabilities/audio/voice-agent · https://docs.x.ai/developers/pricing

Model page: https://postcutoff.com/m/grok-voice-think-fast-2-0/