GrokMCP

grok_chat_completions

Send messages to Grok chat models to generate OpenAI-compatible chat completions, supporting reasoning, vision, and tool calling via the AceDataCloud API.

Instructions

Create a Grok (xAI) chat completion via the AceDataCloud Grok API.

Sends messages to a Grok chat model and returns the generated response in the
OpenAI-compatible chat completion format.

Use this when:
- You want to chat/reason with a Grok model (grok-4 / grok-3 family)
- You need vision/image understanding
- You need tool/function calling with Grok

For generating videos, use grok_text_to_video / grok_image_to_video instead.

Returns:
    JSON response containing the chat completion result.

Input Schema

TableJSON Schema

Name	Required	Description	Default
`n`	No	How many chat completion choices to generate. Default 1.
`seed`	No	Random seed for (best-effort) deterministic sampling.
`stop`	No	Stop sequences where the API will stop generating tokens.
`user`	No	End-user identifier for abuse monitoring.
`audio`	No	Audio output configuration when requesting audio modality.
`model`	No	The Grok chat model: grok-4.5 (default, latest flagship), grok-4 or grok-3.	grok-4
`store`	No	Whether to store the output of this chat completion. Default false.
`tools`	No	List of tools (functions) the model may call.
`top_p`	No	Nucleus sampling probability mass. Default 1.
`stream`	No	Whether to stream partial message deltas. Default False.
`logprobs`	No	Whether to return log probabilities of output tokens.
`messages`	Yes	Conversation messages. Each message is a dict with 'role' ('system'/'user'/'assistant'/'tool') and 'content' keys. Content may be a list of text/image_url parts for image input. Required.
`metadata`	No	Developer-defined metadata attached to the request.
`logit_bias`	No	Token logit bias map.
`max_tokens`	No	Maximum number of tokens to generate.
`modalities`	No	Output modalities requested for this response.
`prediction`	No	Static predicted output content to improve latency.
`temperature`	No	Sampling temperature between 0 and 2. Higher = more random.
`tool_choice`	No	Controls tool calling. 'none', 'auto', 'required', or a dict.
`service_tier`	No	Specifies the processing tier. Options: 'auto' (default), 'default', 'flex' (asynchronous batch-eligible), 'scale', or 'priority'.
`top_logprobs`	No	Number of most likely tokens to return at each token position.
`stream_options`	No	Options for streaming responses.
`response_format`	No	Response format specification (e.g. {"type": "json_object"}).
`presence_penalty`	No	Presence penalty between -2.0 and 2.0. Positive increases topic variety.
`reasoning_effort`	No	Reasoning effort. Only applies to reasoning-capable models; ignored by non-reasoning models.
`frequency_penalty`	No	Frequency penalty between -2.0 and 2.0. Positive decreases repetition.
`web_search_options`	No	Web search configuration for search-capable models.
`parallel_tool_calls`	No	Whether to enable parallel tool calls. Default true.
`max_completion_tokens`	No	Upper bound for tokens generated for a completion.

Output Schema

TableJSON Schema

Name	Required	Description	Default
`result`	Yes

Schema Changelog

Changes observed during successful MCP inspections.

Changed14 schema fields changedv0.1.6Aug 21, 2026

addedInput schema / properties / audio

Added value: +{
+  "anyOf": [
+    {
+      "additionalProperties": true,
+      "type": "object"
+    },
+    {
+      "type": "null"
+    }
+  ],
+  "default": null,
+  "description": "Audio output configuration when requesting audio modality.",
+  "title": "Audio"
+}

addedInput schema / properties / logit_bias

Added value: +{
+  "anyOf": [
+    {
+      "additionalProperties": {
+        "type": "integer"
+      },
+      "type": "object"
+    },
+    {
+      "type": "null"
+    }
+  ],
+  "default": null,
+  "description": "Token logit bias map.",
+  "title": "Logit Bias"
+}

addedInput schema / properties / logprobs

Added value: +{
+  "anyOf": [
+    {
+      "type": "boolean"
+    },
+    {
+      "type": "null"
+    }
+  ],
+  "default": null,
+  "description": "Whether to return log probabilities of output tokens.",
+  "title": "Logprobs"
+}

addedInput schema / properties / max_completion_tokens

Added value: +{
+  "anyOf": [
+    {
+      "type": "integer"
+    },
+    {
+      "type": "null"
+    }
+  ],
+  "default": null,
+  "description": "Upper bound for tokens generated for a completion.",
+  "title": "Max Completion Tokens"
+}

addedInput schema / properties / metadata

Added value: +{
+  "anyOf": [
+    {
+      "additionalProperties": true,
+      "type": "object"
+    },
+    {
+      "type": "null"
+    }
+  ],
+  "default": null,
+  "description": "Developer-defined metadata attached to the request.",
+  "title": "Metadata"
+}

addedInput schema / properties / modalities

Added value: +{
+  "anyOf": [
+    {
+      "items": {
+        "type": "string"
+      },
+      "type": "array"
+    },
+    {
+      "type": "null"
+    }
+  ],
+  "default": null,
+  "description": "Output modalities requested for this response.",
+  "title": "Modalities"
+}

addedInput schema / properties / n

Added value: +{
+  "anyOf": [
+    {
+      "type": "integer"
+    },
+    {
+      "type": "null"
+    }
+  ],
+  "default": null,
+  "description": "How many chat completion choices to generate. Default 1.",
+  "title": "N"
+}

addedInput schema / properties / parallel_tool_calls

Added value: +{
+  "anyOf": [
+    {
+      "type": "boolean"
+    },
+    {
+      "type": "null"
+    }
+  ],
+  "default": null,
+  "description": "Whether to enable parallel tool calls. Default true.",
+  "title": "Parallel Tool Calls"
+}

addedInput schema / properties / prediction

Added value: +{
+  "anyOf": [
+    {
+      "additionalProperties": true,
+      "type": "object"
+    },
+    {
+      "type": "null"
+    }
+  ],
+  "default": null,
+  "description": "Static predicted output content to improve latency.",
+  "title": "Prediction"
+}

addedInput schema / properties / service_tier

Added value: +{
+  "anyOf": [
+    {
+      "enum": [
+        "auto",
+        "default",
+        "flex",
+        "scale",
+        "priority"
+      ],
+      "type": "string"
+    },
+    {
+      "type": "null"
+    }
+  ],
+  "default": null,
+  "description": "Specifies the processing tier. Options: 'auto' (default), 'default', 'flex' (asynchronous batch-eligible), 'scale', or 'priority'.",
+  "title": "Service Tier"
+}

addedInput schema / properties / store

Added value: +{
+  "anyOf": [
+    {
+      "type": "boolean"
+    },
+    {
+      "type": "null"
+    }
+  ],
+  "default": null,
+  "description": "Whether to store the output of this chat completion. Default false.",
+  "title": "Store"
+}

addedInput schema / properties / stream_options

Added value: +{
+  "anyOf": [
+    {
+      "additionalProperties": true,
+      "type": "object"
+    },
+    {
+      "type": "null"
+    }
+  ],
+  "default": null,
+  "description": "Options for streaming responses.",
+  "title": "Stream Options"
+}

addedInput schema / properties / top_logprobs

Added value: +{
+  "anyOf": [
+    {
+      "type": "integer"
+    },
+    {
+      "type": "null"
+    }
+  ],
+  "default": null,
+  "description": "Number of most likely tokens to return at each token position.",
+  "title": "Top Logprobs"
+}

addedInput schema / properties / web_search_options

Added value: +{
+  "anyOf": [
+    {
+      "additionalProperties": true,
+      "type": "object"
+    },
+    {
+      "type": "null"
+    }
+  ],
+  "default": null,
+  "description": "Web search configuration for search-capable models.",
+  "title": "Web Search Options"
+}

Changed5 schema fields changedv0.1.3Jul 27, 2026

changedInput schema / properties / messages / description

Previous value: -"Conversation messages. Each message is a dict with 'role' ('system'/'user'/'assistant'/'tool') and 'content' keys. For vision with grok-2-vision, content may be a list of text/image_url parts. Required."New value: +"Conversation messages. Each message is a dict with 'role' ('system'/'user'/'assistant'/'tool') and 'content' keys. Content may be a list of text/image_url parts for image input. Required."

changedInput schema / properties / model / description

Previous value: -"The Grok chat model. grok-4 (default, flagship) and grok-3 are the broadly available models. Also: grok-4-1-fast, grok-4-1-fast-non-reasoning, grok-3-mini, grok-2-vision (image input) — availability depends on upstream provisioning."New value: +"The Grok chat model: grok-4.5 (default, latest flagship), grok-4 or grok-3."

changedInput schema / properties / model / enum

Previous value: -[
-  "grok-4",
-  "grok-4-1-fast",
-  "grok-4-1-fast-non-reasoning",
-  "grok-3",
-  "grok-3-mini",
-  "grok-2-vision"
-]New value: +[
+  "grok-4.5",
+  "grok-4",
+  "grok-3"
+]

changedInput schema / properties / reasoning_effort / anyOf

Previous value: -[
-  {
-    "enum": [
-      "low",
-      "high"
-    ],
-    "type": "string"
-  },
-  {
-    "type": "null"
-  }
-]New value: +[
+  {
+    "enum": [
+      "minimal",
+      "low",
+      "medium",
+      "high"
+    ],
+    "type": "string"
+  },
+  {
+    "type": "null"
+  }
+]

changedInput schema / properties / reasoning_effort / description

Previous value: -"Reasoning effort: 'low' or 'high'. Only applies to reasoning-capable models (e.g. grok-3-mini). Ignored by non-reasoning models."New value: +"Reasoning effort. Only applies to reasoning-capable models; ignored by non-reasoning models."

First observedv0.1.0Jun 28, 2026

TDQS

A4.3/5.0

Behavior4/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

With no annotations, the description carries the behavioral burden, and it delivers the key traits: it sends messages to a remote Grok model and returns the response in the OpenAI-compatible chat completion format, with vision and function-calling capabilities called out. It does not mention cost, rate limits, or data-sharing implications of calling an external API, but those are inherent to the named API and less critical than the operation/response behavior.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The description is front-loaded and well structured with a short definition, scoped usage bullets, an alternative, and a returns note. Minor redundancy exists between the opening sentence and the 'Returns:' line, but no sentence is wasted.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness4/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a 29-parameter tool with a fully described schema and an output schema, the description covers purpose, use cases, exclusions, and response format. It is not a 5 because it omits an example of the required messages structure and model-specific caveats, though the schema compensates for most of this.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema description coverage is 100%, so the input schema already documents all 29 parameters. The narrative mentions vision and tool calling, which aligns with messages/tools, but it does not add syntax or format details beyond what the schema provides; baseline 3 is appropriate.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description opens with a specific action and resource: 'Create a Grok (xAI) chat completion via the AceDataCloud Grok API' and then clarifies it sends messages to a Grok model and returns an OpenAI-compatible response. It also separates itself from sibling generation tools by telling agents to use grok_text_to_video / grok_image_to_video for video.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines5/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

It provides explicit 'Use this when' bullets for chat/reasoning, vision, and tool calling, and an explicit exclusion: 'For generating videos, use grok_text_to_video / grok_image_to_video instead.' This tells the agent both when to invoke this tool and when not to.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Deploy Server