Documentation 2026-10-01 23:57 UTC to 2026-10-02 23:57 UTC

8 8 

9### LiteLLM9### LiteLLM

10 10 

11LiteLLM provides a simple SDK or proxy server for calling different LLM providers. If you're using LiteLLM, integrating xAI as your provider is straightforward—just swap out the model name and API key to xAI's Grok model in your configuration.11LiteLLM provides a simple SDK or proxy server for calling different LLM providers. If you're using LiteLLM, integrating SpaceXAI as your provider is straightforward—just swap out the model name and API key to SpaceXAI's Grok model in your configuration.

12 12 

13For latest information and more examples, visit [LiteLLM xAI Provider Documentation](https://docs.litellm.ai/docs/providers/xai).13For latest information and more examples, visit [LiteLLM xAI Provider Documentation](https://docs.litellm.ai/docs/providers/xai).

14 14 

2 2 

3# Google Cloud Vertex AI3# Google Cloud Vertex AI

4 4 

5Access xAI’s Grok models through Google Cloud’s managed platform with enterprise security, governance, and unified billing.5Access SpaceXAI’s Grok models through Google Cloud’s managed platform with enterprise security, governance, and unified billing.

6 6 

7This guide walks through setting up and using Grok models on Google Cloud Vertex AI / Gemini Enterprise Agent Platform. Grok on Vertex AI is accessed as a partner model through the OpenAI-compatible API, including the Responses API and Chat Completions. Models are enabled through Model Garden.7This guide walks through setting up and using Grok models on Google Cloud Vertex AI / Gemini Enterprise Agent Platform. Grok on Vertex AI is accessed as a partner model through the OpenAI-compatible API, including the Responses API and Chat Completions. Models are enabled through Model Garden.

8 8 

181* Explore enabled models in Model Garden.181* Explore enabled models in Model Garden.

182* Build agentic applications that use Grok’s tool-calling strengths.182* Build agentic applications that use Grok’s tool-calling strengths.

183* Integrate with Google Cloud services such as Cloud Functions and Vertex AI Pipelines.183* Integrate with Google Cloud services such as Cloud Functions and Vertex AI Pipelines.

184* Review the full xAI Grok documentation and model cards for prompting tips and capabilities.184* Review the full SpaceXAI Grok documentation and model cards for prompting tips and capabilities.

2 2 

3# Microsoft Foundry3# Microsoft Foundry

4 4 

5Access xAI’s frontier reasoning and agentic models through Azure AI Foundry with enterprise-grade security, governance, and unified billing.5Access SpaceXAI’s frontier reasoning and agentic models through Azure AI Foundry with enterprise-grade security, governance, and unified billing.

6 6 

7This guide walks through setting up and using Grok models on Microsoft Foundry. Grok models on Foundry give you strong reasoning, native tool use, enterprise authentication through Microsoft Entra ID, Azure-native monitoring, and an OpenAI-compatible API.7This guide walks through setting up and using Grok models on Microsoft Foundry. Grok models on Foundry give you strong reasoning, native tool use, enterprise authentication through Microsoft Entra ID, Azure-native monitoring, and an OpenAI-compatible API.

8 8 

9Usage is billed through the Azure Marketplace / your Azure subscription. Grok models are delivered through the xAI–Microsoft partnership with Azure-managed endpoints and optional Azure AI Content Safety layers. Review the specific model card in the Foundry catalog for the latest details on data processing, retention, and terms.9Usage is billed through the Azure Marketplace / your Azure subscription. Grok models are delivered through the SpaceXAI–Microsoft partnership with Azure-managed endpoints and optional Azure AI Content Safety layers. Review the specific model card in the Foundry catalog for the latest details on data processing, retention, and terms.

10 10 

11Grok on Foundry works with the official OpenAI Python/TypeScript SDKs, `azure-ai-projects`, LangChain, Semantic Kernel, LlamaIndex, and most OpenAI-compatible frameworks. Streaming, tool calling, and structured outputs are supported.11Grok on Foundry works with the official OpenAI Python/TypeScript SDKs, `azure-ai-projects`, LangChain, Semantic Kernel, LlamaIndex, and most OpenAI-compatible frameworks. Streaming, tool calling, and structured outputs are supported.

12 12 

216 216 

217## Correlation IDs and debugging217## Correlation IDs and debugging

218 218 

219Foundry includes standard Azure request identifiers in response headers, such as `request-id`, `apim-request-id`, and `x-ms-request-id`. When contacting Microsoft or xAI support, include these IDs with your deployment name and approximate timestamp.219Foundry includes standard Azure request identifiers in response headers, such as `request-id`, `apim-request-id`, and `x-ms-request-id`. When contacting Microsoft or SpaceXAI support, include these IDs with your deployment name and approximate timestamp.

220 220 

221## Feature support and capabilities221## Feature support and capabilities

222 222 

231 231 

232## Safety and responsible AI232## Safety and responsible AI

233 233 

234Grok models include xAI’s safety training and alignment. On Foundry, Azure AI Content Safety is available and often enabled by default or easily integrated.234Grok models include SpaceXAI’s safety training and alignment. On Foundry, Azure AI Content Safety is available and often enabled by default or easily integrated.

235 235 

236Before production deployment:236Before production deployment:

237 237 

247* Validate vision/multimodal support and exact parameter availability for your chosen model/deployment.247* Validate vision/multimodal support and exact parameter availability for your chosen model/deployment.

248* Rate limits and quotas are managed at the Azure resource level.248* Rate limits and quotas are managed at the Azure resource level.

249 249 

250For the authoritative list of supported parameters and behaviors, consult the model card inside Azure AI Foundry and xAI Grok documentation linked from the catalog.250For the authoritative list of supported parameters and behaviors, consult the model card inside Azure AI Foundry and SpaceXAI Grok documentation linked from the catalog.

251 251 

252## Best practices for production252## Best practices for production

253 253 

43 43 

44You can visit [xAI Accounts](https://accounts.x.ai) to manage your account.44You can visit [xAI Accounts](https://accounts.x.ai) to manage your account.

45 45 

46Please note the xAI account is different from the X account, and xAI cannot assist you with X account issues. Please46Please note the xAI account is different from the X account, and SpaceXAI cannot assist you with X account issues. Please

47contact X via [X Help Center](https://help.x.com/) or Premium Support if you encounter any issues with your X account.47contact X via [X Help Center](https://help.x.com/) or Premium Support if you encounter any issues with your X account.

48 48 

49## I received an email of someone logging into my xAI account49## I received an email of someone logging into my xAI account

50 50 

51xAI will send an email to you when someone logs into your xAI account. The login location is an approximation based on your IP address, which is dependent on your network setup and ISP and might not reflect exactly where the login happened.51SpaceXAI will send an email to you when someone logs into your xAI account. The login location is an approximation based on your IP address, which is dependent on your network setup and ISP and might not reflect exactly where the login happened.

52 52 

53If you think the login is not you, please [reset your password](https://accounts.x.ai/request-reset-password) and [clear your login sessions](https://accounts.x.ai/sessions). We also recommend all users to [add a multi-factor authentication method](https://accounts.x.ai/security).53If you think the login is not you, please [reset your password](https://accounts.x.ai/request-reset-password) and [clear your login sessions](https://accounts.x.ai/sessions). We also recommend all users to [add a multi-factor authentication method](https://accounts.x.ai/security).

54 54 

58 58 

59You can visit [xAI Accounts](https://accounts.x.ai/account) to delete your account. You can restore your account by logging in again and confirming restoration within 30 days.59You can visit [xAI Accounts](https://accounts.x.ai/account) to delete your account. You can restore your account by logging in again and confirming restoration within 30 days.

60 60 

61You can cancel the deletion within 30 days by logging in again to any xAI websites and following the prompt to confirm restoring the account.61You can cancel the deletion within 30 days by logging in again to any SpaceXAI websites and following the prompt to confirm restoring the account.

62 62 

63For privacy requests, please go to: https://privacy.x.ai.63For privacy requests, please go to: https://privacy.x.ai.

20 20 

21Please refer to our [Legal Resources](https://x.ai/legal) for our Enterprise Terms of Service and Data Processing Addendum.21Please refer to our [Legal Resources](https://x.ai/legal) for our Enterprise Terms of Service and Data Processing Addendum.

22 22 

23### Does xAI sell crypto tokens?23### Does SpaceXAI sell crypto tokens?

24 24 

25xAI is not affiliated with any cryptocurrency. We are aware of several scam websites that unlawfully use our name and logo.25SpaceXAI is not affiliated with any cryptocurrency. We are aware of several scam websites that unlawfully use our name and logo.

4 4 

5## What are teams?5## What are teams?

6 6 

7Teams are the level at which xAI tracks API usage, processes billing, and issues invoices.7Teams are the level at which SpaceXAI tracks API usage, processes billing, and issues invoices.

8 8 

9* If you’re the team creator and don’t need a new team, you can rename your Personal Team and add members instead of creating a new one.9* If you’re the team creator and don’t need a new team, you can rename your Personal Team and add members instead of creating a new one.

10* Each team has **roles**:10* Each team has **roles**:

14 14 

15## Which team am I on?15## Which team am I on?

16 16 

17When you sign up for xAI, you’re automatically assigned to a **Personal Team**, which you can view the top bar of [xAI Console](https://console.x.ai?utm_source=docs\&utm_medium=referral\&utm_campaign=developers-faq-team-management\&utm_content=console-home).17When you sign up for SpaceXAI, you’re automatically assigned to a **Personal Team**, which you can view the top bar of [xAI Console](https://console.x.ai?utm_source=docs\&utm_medium=referral\&utm_campaign=developers-faq-team-management\&utm_content=console-home).

18 18 

19## How can I manage teams and team members?19## How can I manage teams and team members?

20 20 

97 97 

98### Get Started with Our Tester Apps98### Get Started with Our Tester Apps

99 99 

100* **[iOS Tester App](https://github.com/xai-org/xai-cookbook/tree/main/iOS/VoiceTesterApp)** — A Swift-based iOS app to act as a guide for setting up voice agents in your apps.100* **[iOS Tester App](https://github.com/xai-org/xai-cookbook/tree/main/examples/voice-agent-mobile/swift)** — A Swift-based iOS app to act as a guide for setting up voice agents in your apps.

101* **[Web Agent (WebSocket)](https://github.com/xai-org/xai-cookbook/tree/main/voice-examples/agent/web)** — A web app voice agent using WebSocket.101* **[Web Agent (WebSocket)](https://github.com/xai-org/xai-cookbook/tree/main/examples/voice-agent-web)** — A web app voice agent using WebSocket.

102* **[WebRTC Agent](https://github.com/xai-org/xai-cookbook/tree/main/voice-examples/agent/webrtc)** — A web app voice agent using WebRTC.102* **[WebRTC Agent](https://github.com/xai-org/xai-cookbook/tree/main/examples/voice-agent-webrtc)** — A web app voice agent using WebRTC.

103* **[Telephony Agent](https://github.com/xai-org/xai-cookbook/tree/main/voice-examples/agent/telephony)** — A callable phone agent using Twilio.103* **[Telephony Agent](https://github.com/xai-org/xai-cookbook/tree/main/examples/voice-agent-phone)** — A callable phone agent using Twilio.

104 104 

105## Authentication105## Authentication

106 106 

147| `voice` | string | Voice selection: any built-in voice (e.g. `eve`) or a [custom voice ID](/developers/model-capabilities/audio/custom-voices) (see [Available Voices](#available-voices)) |147| `voice` | string | Voice selection: any built-in voice (e.g. `eve`) or a [custom voice ID](/developers/model-capabilities/audio/custom-voices) (see [Available Voices](#available-voices)) |

148| `tools` | array | Tools available to the voice agent. Supports `file_search`, `web_search`, `x_search`, `mcp`, and `function` types. See [Using Tools](#using-tools-with-grok-speech-to-speech-api). |148| `tools` | array | Tools available to the voice agent. Supports `file_search`, `web_search`, `x_search`, `mcp`, and `function` types. See [Using Tools](#using-tools-with-grok-speech-to-speech-api). |

149| `turn_detection.type` | string | null | `"server_vad"` for automatic detection, `null` for manual text turns |149| `turn_detection.type` | string | null | `"server_vad"` for automatic detection, `null` for manual text turns |

150| `turn_detection.threshold` | number | optional | VAD activation threshold (0.1–0.9). Higher values require louder audio to trigger. Default: `0.85`. |

151| `turn_detection.silence_duration_ms` | number | optional | How long the user must be silent (in ms) before the server ends the turn (0–10000). Higher values let users pause longer without being cut off. |

152| `turn_detection.prefix_padding_ms` | number | optional | Amount of audio (in ms) to include before the detected start of speech (0–10000). Helps capture the beginning of words that might otherwise be clipped by the VAD. Default: `333`. |

153| `turn_detection.idle_timeout_ms` | number | optional | When set, the server proactively re-engages the user if no speech is detected for this many milliseconds after the assistant finishes responding. The timer re-arms after every response, so it fires repeatedly each `idle_timeout_ms` until the user speaks. Default: `null`. |150| `turn_detection.idle_timeout_ms` | number | optional | When set, the server proactively re-engages the user if no speech is detected for this many milliseconds after the assistant finishes responding. The timer re-arms after every response, so it fires repeatedly each `idle_timeout_ms` until the user speaks. Default: `null`. |

154| `resumption.enabled` | boolean | optional | Opt in to [Session Resumption](#session-resumption): the server caches conversation turns keyed by `conversation_id` and replays them on reconnect so the model stays conditioned on prior context. Defaults to `false`. See [Session Resumption](#session-resumption). |151| `resumption.enabled` | boolean | optional | Opt in to [Session Resumption](#session-resumption): the server caches conversation turns keyed by `conversation_id` and replays them on reconnect so the model stays conditioned on prior context. Defaults to `false`. See [Session Resumption](#session-resumption). |

155| `audio.input.format.type` | string | Input codec: `"audio/pcm"`, `"audio/pcmu"`, `"audio/pcma"`, or `"audio/opus"` |152| `audio.input.format.type` | string | Input codec: `"audio/pcm"`, `"audio/pcmu"`, `"audio/pcma"`, or `"audio/opus"` |

83| Model | Description |83| Model | Description |

84|-------|-------------|84|-------|-------------|

85| `grok-voice-transcribe-2.0` | Our best transcription model. Default when `model` is omitted. |85| `grok-voice-transcribe-2.0` | Our best transcription model. Default when `model` is omitted. |

86| `grok-voice-transcribe-1.0` | Original model. Pin this slug to keep it. |86| `grok-voice-transcribe-1.0` | Original model. Deprecated; end of life October 2, 2026. Requests are routed to `grok-voice-transcribe-2.0` at the same price, with higher accuracy. |

87 87 

88## Supported Languages88## Supported Languages

89 89 

113|-----------|------|---------|----------|-------------|113|-----------|------|---------|----------|-------------|

114| `file` | file | | ✓† | Audio file to transcribe. Max **500 MB**. See [Supported Formats](#supported-audio-formats). Must be the last field in the multipart form. |114| `file` | file | | ✓† | Audio file to transcribe. Max **500 MB**. See [Supported Formats](#supported-audio-formats). Must be the last field in the multipart form. |

115| `url` | string | | ✓† | URL of an audio file to download and transcribe (server-side). |115| `url` | string | | ✓† | URL of an audio file to download and transcribe (server-side). |

116| `model` | string | `grok-voice-transcribe-2.0` | | `grok-voice-transcribe-1.0` or `grok-voice-transcribe-2.0`. |116| `model` | string | `grok-voice-transcribe-2.0` | | `grok-voice-transcribe-2.0` (default) or `grok-voice-transcribe-1.0` (deprecated; routed to 2.0). |

117| `audio_format` | string | | | Format hint for raw/headerless audio: `pcm`, `mulaw`, `alaw`. Container formats are auto-detected — do not set this field for MP3, WAV, etc. |117| `audio_format` | string | | | Format hint for raw/headerless audio: `pcm`, `mulaw`, `alaw`. Container formats are auto-detected — do not set this field for MP3, WAV, etc. |

118| `sample_rate` | integer | | | Sample rate in Hz. Only required for raw audio (`pcm`, `mulaw`, `alaw`). Supported: `8000`, `16000`, `22050`, `24000`, `44100`, `48000`. |118| `sample_rate` | integer | | | Sample rate in Hz. Only required for raw audio (`pcm`, `mulaw`, `alaw`). Supported: `8000`, `16000`, `22050`, `24000`, `44100`, `48000`. |

119| `language` | string | | | Language code (e.g. `en`, `fr`, `de`). Used with `format=true` to enable text formatting. See [Supported Languages](#supported-languages). |119| `language` | string | | | Language code (e.g. `en`, `fr`, `de`). Used with `format=true` to enable text formatting. See [Supported Languages](#supported-languages). |

220| `interim_results` | boolean | `false` | When `true`, emit partial transcripts `is_final=false` every ~500 ms. |220| `interim_results` | boolean | `false` | When `true`, emit partial transcripts `is_final=false` every ~500 ms. |

221| `endpointing` | integer | `400` | Silence duration (ms) before utterance-final event. Range: 0–5000. `0` = fire on any VAD silence boundary. |221| `endpointing` | integer | `400` | Silence duration (ms) before utterance-final event. Range: 0–5000. `0` = fire on any VAD silence boundary. |

222| `language` | string | | Language code for text formatting. See [Supported Languages](#supported-languages). |222| `language` | string | | Language code for text formatting. See [Supported Languages](#supported-languages). |

223| `model` | string | `grok-voice-transcribe-2.0` | `grok-voice-transcribe-1.0` or `grok-voice-transcribe-2.0`. |223| `model` | string | `grok-voice-transcribe-2.0` | `grok-voice-transcribe-2.0` (default) or `grok-voice-transcribe-1.0` (deprecated; routed to 2.0). |

224| `diarize` | boolean | | When `true`, enables speaker diarization. Words include a `speaker` field identifying the detected speaker. |224| `diarize` | boolean | | When `true`, enables speaker diarization. Words include a `speaker` field identifying the detected speaker. |

225| `filler_words` | boolean | `false` | When `true`, filler words (e.g. `uh`, `um`, `er`) are included in the transcript. When `false` (default), filler words are automatically removed. |225| `filler_words` | boolean | `false` | When `true`, filler words (e.g. `uh`, `um`, `er`) are included in the transcript. When `false` (default), filler words are automatically removed. |

226| `multichannel` | boolean | `false` | Per-channel transcription. Requires `channels` ≥ 2. Not supported with `encoding=opus`. |226| `multichannel` | boolean | `false` | Per-channel transcription. Requires `channels` ≥ 2. Not supported with `encoding=opus`. |

69 69 

70```70```

71 71 

72**Demo Apps:** [Web Agent](https://github.com/xai-org/xai-cookbook/tree/main/voice-examples/agent/web) · [Twilio Phone Agent](https://github.com/xai-org/xai-cookbook/tree/main/voice-examples/agent/telephony) · [WebRTC Agent](https://github.com/xai-org/xai-cookbook/tree/main/voice-examples/agent/webrtc) · [iOS Tester App](https://github.com/xai-org/xai-cookbook/tree/main/iOS/VoiceTesterApp)72**Demo Apps:** [Web Agent](https://github.com/xai-org/xai-cookbook/tree/main/examples/voice-agent-web) · [Twilio Phone Agent](https://github.com/xai-org/xai-cookbook/tree/main/examples/voice-agent-phone) · [WebRTC Agent](https://github.com/xai-org/xai-cookbook/tree/main/examples/voice-agent-webrtc) · [iOS Tester App](https://github.com/xai-org/xai-cookbook/tree/main/examples/voice-agent-mobile/swift)

73 73 

74## Text to Speech74## Text to Speech

75 75 

132 132 

133## Speech to Text133## Speech to Text

134 134 

135Transcribe audio files in a single call or stream over WebSocket. Use `grok-voice-transcribe-1.0` or `grok-voice-transcribe-2.0`; the default is `grok-voice-transcribe-2.0`. 12 audio formats, word-level timestamps, multichannel, speaker diarization, Smart Turn end-of-turn detection, and 25 languages.135Transcribe audio files in a single call or stream over WebSocket. The default is `grok-voice-transcribe-2.0`. `grok-voice-transcribe-1.0` is deprecated and reaches end of life on October 2, 2026; requests to that slug are routed to 2.0. 12 audio formats, word-level timestamps, multichannel, speaker diarization, Smart Turn end-of-turn detection, and 25 languages.

136 136 

137```bash137```bash

138curl -X POST https://api.x.ai/v1/stt \138curl -X POST https://api.x.ai/v1/stt \

42| grok-build-0.1 | T0: 37, T1: 50, T2: 75, T3: 125, T4: 208 | T0: 10M, T1: 15M, T2: 25M, T3: 45M, T4: 85M |42| grok-build-0.1 | T0: 37, T1: 50, T2: 75, T3: 125, T4: 208 | T0: 10M, T1: 15M, T2: 25M, T3: 45M, T4: 85M |

43| grok-4.20-multi-agent-0309 | T0: 9, T1: 12, T2: 18, T3: 31, T4: 56 | T0: 2.5M, T1: 3.7M, T2: 6.2M, T3: 11M, T4: 21M |43| grok-4.20-multi-agent-0309 | T0: 9, T1: 12, T2: 18, T3: 31, T4: 56 | T0: 2.5M, T1: 3.7M, T2: 6.2M, T3: 11M, T4: 21M |

44| grok-imagine-image | T0: 6, T1: 12, T2: 25, T3: 50, T4: 100 | — |44| grok-imagine-image | T0: 6, T1: 12, T2: 25, T3: 50, T4: 100 | — |

45| grok-imagine-image-2.0 | T0: 6, T1: 12, T2: 25, T3: 50, T4: 100 | — |

46| grok-imagine-image-quality | T0: 6, T1: 12, T2: 25, T3: 50, T4: 100 | — |45| grok-imagine-image-quality | T0: 6, T1: 12, T2: 25, T3: 50, T4: 100 | — |

46| grok-imagine-image-2.0 | T0: 6, T1: 12, T2: 25, T3: 50, T4: 100 | — |

47| grok-imagine-video-1.5-lite | T0: 10, T1: 20, T2: 39, T3: 79, T4: 158 | — |47| grok-imagine-video-1.5-lite | T0: 10, T1: 20, T2: 39, T3: 79, T4: 158 | — |

48| grok-imagine-video-1.5 | T0: 10, T1: 20, T2: 39, T3: 79, T4: 158 | — |48| grok-imagine-video-1.5 | T0: 10, T1: 20, T2: 39, T3: 79, T4: 158 | — |

49| grok-imagine-video | T0: 10, T1: 20, T2: 39, T3: 79, T4: 158 | — |49| grok-imagine-video | T0: 10, T1: 20, T2: 39, T3: 79, T4: 158 | — |

26 26 

27* `logprobs` (boolean | null) — Whether to return log probabilities of the output tokens or not. If true, returns the log probabilities of each output token returned in the content of message. Not supported by models \`grok-4.20\` and newer; the field will be silently ignored if set.27* `logprobs` (boolean | null) — Whether to return log probabilities of the output tokens or not. If true, returns the log probabilities of each output token returned in the content of message. Not supported by models \`grok-4.20\` and newer; the field will be silently ignored if set.

28 28 

29* `max_output_tokens` (integer | null) — Max number of tokens that can be generated in a response. This includes both output and reasoning tokens. Defaults to 128,000 when unset; set a larger value to allow longer generations.29* `max_output_tokens` (integer | null) — Max number of tokens that can be generated in a response. Only applies to visible output tokens (i.e. does not apply to tokens used for reasoning or function calls). Defaults to 128,000 when unset; set a larger value to allow longer generations.

30 30 

31* `max_turns` (integer | null) — Maximum number of agentic tool calling turns allowed for this request.31* `max_turns` (integer | null) — Maximum number of agentic tool calling turns allowed for this request.

32 If not set, defaults to the server's global cap.32 If not set, defaults to the server's global cap.

123 123 

124* `instructions` (string | null) — A system (or developer) message inserted into the model's context.124* `instructions` (string | null) — A system (or developer) message inserted into the model's context.

125 125 

126* `max_output_tokens` (integer | null) — Max number of tokens that can be generated in a response. This includes both output and reasoning tokens.126* `max_output_tokens` (integer | null) — Max number of tokens that can be generated in a response. Only applies to visible output tokens (i.e. does not apply to tokens used for reasoning or function calls).

127 127 

128* `max_tool_calls` (integer | null) — The maximum number of tool calls allowed for this response.128* `max_tool_calls` (integer | null) — The maximum number of tool calls allowed for this response.

129 129 

383 383 

384* `instructions` (string | null) — A system (or developer) message inserted into the model's context.384* `instructions` (string | null) — A system (or developer) message inserted into the model's context.

385 385 

386* `max_output_tokens` (integer | null) — Max number of tokens that can be generated in a response. This includes both output and reasoning tokens.386* `max_output_tokens` (integer | null) — Max number of tokens that can be generated in a response. Only applies to visible output tokens (i.e. does not apply to tokens used for reasoning or function calls).

387 387 

388* `max_tool_calls` (integer | null) — The maximum number of tool calls allowed for this response.388* `max_tool_calls` (integer | null) — The maximum number of tool calls allowed for this response.

389 389 

140 140 

141WebSocket endpoint: `wss://api.x.ai/v1/stt`141WebSocket endpoint: `wss://api.x.ai/v1/stt`

142 142 

143Real-time streaming speech-to-text via WebSocket. Stream raw audio as binary frames and receive JSON transcript events as the audio is processed. Configuration is done via query parameters at connection time. Use grok-voice-transcribe-1.0 or grok-voice-transcribe-2.0; the default is grok-voice-transcribe-2.0.143Real-time streaming speech-to-text via WebSocket. Stream raw audio as binary frames and receive JSON transcript events as the audio is processed. Configuration is done via query parameters at connection time. The default is grok-voice-transcribe-2.0. grok-voice-transcribe-1.0 is deprecated and reaches end of life on October 2, 2026; requests to that slug are routed to grok-voice-transcribe-2.0.

144 144 

145Full schemas and examples: [`/stt-streaming.ws.json`](/stt-streaming.ws.json)145Full schemas and examples: [`/stt-streaming.ws.json`](/stt-streaming.ws.json)

146 146 

156 156 

157* `language` (string, optional, default: ) — Language code (e.g. \`en\`, \`fr\`, \`de\`, \`ja\`). When set, enables Inverse Text Normalization — spoken-form numbers, currencies, and units are converted to their written form.157* `language` (string, optional, default: ) — Language code (e.g. \`en\`, \`fr\`, \`de\`, \`ja\`). When set, enables Inverse Text Normalization — spoken-form numbers, currencies, and units are converted to their written form.

158 158 

159* `model` (string, optional, default: grok-voice-transcribe-2.0) — \`grok-voice-transcribe-1.0\` or \`grok-voice-transcribe-2.0\`. Defaults to \`grok-voice-transcribe-2.0\`.159* `model` (string, optional, default: grok-voice-transcribe-2.0) — \`grok-voice-transcribe-2.0\` (default) or \`grok-voice-transcribe-1.0\` (deprecated; routed to 2.0 as of October 2, 2026).

160 160 

161* `multichannel` (boolean, optional, default: false) — When \`true\`, enables per-channel transcription for interleaved multichannel audio. Requires \`channels\` to be set to ≥ 2. Not supported with \`encoding=opus\`.161* `multichannel` (boolean, optional, default: false) — When \`true\`, enables per-channel transcription for interleaved multichannel audio. Requires \`channels\` to be set to ≥ 2. Not supported with \`encoding=opus\`.

162 162 

1138 1138 

1139WebSocket endpoint: `wss://api.x.ai/v1/stt`1139WebSocket endpoint: `wss://api.x.ai/v1/stt`

1140 1140 

1141Real-time streaming speech-to-text via WebSocket. Stream raw audio as binary frames and receive JSON transcript events as the audio is processed. Configuration is done via query parameters at connection time. Use grok-voice-transcribe-1.0 or grok-voice-transcribe-2.0; the default is grok-voice-transcribe-2.0.1141Real-time streaming speech-to-text via WebSocket. Stream raw audio as binary frames and receive JSON transcript events as the audio is processed. Configuration is done via query parameters at connection time. The default is grok-voice-transcribe-2.0. grok-voice-transcribe-1.0 is deprecated and reaches end of life on October 2, 2026; requests to that slug are routed to grok-voice-transcribe-2.0.

1142 1142 

1143Full schemas and examples: [`/stt-streaming.ws.json`](/stt-streaming.ws.json)1143Full schemas and examples: [`/stt-streaming.ws.json`](/stt-streaming.ws.json)

1144 1144 

1154 1154 

1155* `language` (string, optional, default: ) — Language code (e.g. \`en\`, \`fr\`, \`de\`, \`ja\`). When set, enables Inverse Text Normalization — spoken-form numbers, currencies, and units are converted to their written form.1155* `language` (string, optional, default: ) — Language code (e.g. \`en\`, \`fr\`, \`de\`, \`ja\`). When set, enables Inverse Text Normalization — spoken-form numbers, currencies, and units are converted to their written form.

1156 1156 

1157* `model` (string, optional, default: grok-voice-transcribe-2.0) — \`grok-voice-transcribe-1.0\` or \`grok-voice-transcribe-2.0\`. Defaults to \`grok-voice-transcribe-2.0\`.1157* `model` (string, optional, default: grok-voice-transcribe-2.0) — \`grok-voice-transcribe-2.0\` (default) or \`grok-voice-transcribe-1.0\` (deprecated; routed to 2.0 as of October 2, 2026).

1158 1158 

1159* `multichannel` (boolean, optional, default: false) — When \`true\`, enables per-channel transcription for interleaved multichannel audio. Requires \`channels\` to be set to ≥ 2. Not supported with \`encoding=opus\`.1159* `multichannel` (boolean, optional, default: false) — When \`true\`, enables per-channel transcription for interleaved multichannel audio. Requires \`channels\` to be set to ≥ 2. Not supported with \`encoding=opus\`.

1160 1160 

6need a [management key](https://console.x.ai/team/default/management-keys?utm_source=docs\&utm_medium=referral\&utm_campaign=developers-rest-api-reference-management\&utm_content=management-keys) in6need a [management key](https://console.x.ai/team/default/management-keys?utm_source=docs\&utm_medium=referral\&utm_campaign=developers-rest-api-reference-management\&utm_content=management-keys) in

7order to use this API. The base URL for all endpoints is `https://management-api.x.ai`.7order to use this API. The base URL for all endpoints is `https://management-api.x.ai`.

8 8 

9The Management API serves as a dedicated interface to the xAI platform, empowering developers and teams to9The Management API serves as a dedicated interface to the SpaceXAI platform, empowering developers and teams to

10programmatically manage their xAI API teams.10programmatically manage their xAI API teams.

11 11 

12For example, users can provision their API key, handle access controls,12For example, users can provision their API key, handle access controls,