Documentation 2026-07-18 00:00 UTC to 2026-07-21 20:58 UTC

4 4 

5Grok 4.5 is SpaceXAI's frontier model built for coding, agentic tasks, and knowledge work. It was trained in SpaceXAI's data centers in Memphis with new datasets spanning science, engineering, and math.5Grok 4.5 is SpaceXAI's frontier model built for coding, agentic tasks, and knowledge work. It was trained in SpaceXAI's data centers in Memphis with new datasets spanning science, engineering, and math.

6 6 

7> Grok 4.5 isn't available in the API console for EU users yet. Availability is expected later this month.

8 

9## Using the API7## Using the API

10 8 

11If you already have an [API key](https://console.x.ai/team/default/api-keys), set the model name to `grok-4.5`:9If you already have an [API key](https://console.x.ai/team/default/api-keys), set the model name to `grok-4.5`:

41| 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 |41| 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 |

42| grok-imagine-image-quality | 5 | — |42| grok-imagine-image-quality | 5 | — |

43| grok-imagine-image | 5 | — |43| grok-imagine-image | 5 | — |

44| grok-imagine-video-1.5 | 10 | — |

45| grok-imagine-video | 10 | — |44| grok-imagine-video | 10 | — |

45| grok-imagine-video-1.5 | 10 | — |

46 46 

47### What counts toward TPM47### What counts toward TPM

48 48 

16 16 

17* `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.17* `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.

18 18 

19* `max_completion_tokens` (integer | null) — An upper bound for the number of tokens that can be generated for a completion, only applies to visible output tokens (i.e. does not apply to tokens used for reasoning or function calls). Defaults to None, meaning the model will generate as many tokens as needed up until the model's maximum context length.19* `max_completion_tokens` (integer | null) — An upper bound for the number of tokens that can be generated for a completion, 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.

20 20 

21* `max_tokens` (integer | null) — \\\[DEPRECATED\\] The maximum number of tokens that can be generated in the chat completion. Deprecated in favor of \`max\_completion\_tokens\`.21* `max_tokens` (integer | null) — \\\[DEPRECATED\\] The maximum number of tokens that can be generated in the chat completion. Deprecated in favor of \`max\_completion\_tokens\`.

22 22 

245 245 

246* `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.246* `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.

247 247 

248* `max_output_tokens` (integer | null) — Max number of tokens that can be generated in a response. This includes both output and reasoning tokens.248* `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.

249 249 

250* `max_turns` (integer | null) — Maximum number of agentic tool calling turns allowed for this request.250* `max_turns` (integer | null) — Maximum number of agentic tool calling turns allowed for this request.

251 If not set, defaults to the server's global cap.251 If not set, defaults to the server's global cap.