Home / Servers / DigitalOcean API / inference_create_messages
Create the next assistant message
An MCP tool from the DigitalOcean API server, hosted on mcp.hjlabs.in.
Create the next assistant message Send a structured list of input messages with text and/or image content, and the model will generate the next message in the conversation. [POST /v1/messages]
Name on the gateway: digitalocean_api__inference_create_messages · may change things.
Parameters
| Name | Type | Description |
|---|---|---|
model required | string | Model ID (for example claude-opus-4-6 or a serverless model id). |
max_tokens required | integer | Maximum tokens to generate before stopping. |
messages required | array | Conversation turns. Each item has role user or assistant and content as a… |
system | string | System prompt as plain text or as an array of text blocks. |
stop_sequences | array | Custom strings that stop generation when produced. |
stream | boolean | When true, the response is streamed using server-sent events (SSE). |
temperature | number | Sampling temperature between 0.0 and 1.0. |
top_p | number | Nucleus sampling; use either temperature or top_p, not both. |
top_k | integer | Top-K sampling cutoff. |
tools | array | Tool definitions the model may invoke. |
tool_choice | object | Controls how the model uses tools: automatic selection, require any tool… |
metadata | object | Optional request metadata. |
reasoning_effort | none | minimal | low | medium | high | xhigh | DigitalOcean extension for reasoning-capable models. Ignored by executors that… |
speed | standard | fast | DigitalOcean extension for preferred inference speed. Ignored when not… |
thinking | object | Extended thinking configuration. Executors that do not support thinking may… |
How to call it
Through the Smart Gateway, any client can run it by name with call_tool, or let the gateway find it with search_tools:
{
"name": "digitalocean_api__inference_create_messages",
"arguments": {
"model": "…",
"max_tokens": 0,
"messages": []
}
}Or connect the DigitalOcean API server directly, which lists 874 tools including this one:
claude mcp add --transport http hj-digitalocean-api https://mcp.hjlabs.in/mcp/digitalocean-api \
--header "Authorization: Bearer YOUR_API_KEY"Try asking
Use the create the next assistant message tool with model set to ….