Create a Message¶

Generate a reply using the Anthropic Messages format and receive text or thinking blocks.

POST https://token.moi.matrixorigin.cn/v1/messages

Before you begin¶

Prepare a personal access token or service account API key for Genesis. See Access credentials for credential selection and permissions.
Use List callable models to select a model that supports Messages.

Request body¶

Replace $GENESIS_ACCESS_TOKEN and $MODEL_ID with your credential and the selected model ID.

Request example

curl -X POST \
  "https://token.moi.matrixorigin.cn/v1/messages" \
  -H "Authorization: Bearer $GENESIS_ACCESS_TOKEN" \
  -H 'Content-Type: application/json' \
  -d '{
  "model": "'"$MODEL_ID"'",
  "max_tokens": 256,
  "messages": [
    {
      "role": "user",
      "content": [
        {
          "type": "text",
          "text": "Reply with OK."
        }
      ]
    }
  ],
  "stream": false
}'

Parameters

The fields below cover text messages. Images, documents, tools, and explicit thinking configuration require corresponding model capabilities.

Parameter	Type	Required	Description
`model`	string	Yes	Selected model ID from `data[].id` in the model list.
`max_tokens`	integer	Yes	Maximum generated tokens; use a positive integer.
`messages`	array of object	Yes	Nonempty messages in conversation order.
`messages[].role`	string	Yes	Message role: `user` or `assistant` for text conversations.
`messages[].content`	string or array of object	Yes	Text or text blocks.
`messages[].content[].type`	string	Required for a text block	Use `text` for a text block.
`messages[].content[].text`	string	Required for a text block	Nonempty text content.
`stream`	boolean	No	Set to `true` for Messages SSE events.

Successful response¶

A non-streaming request returns the generated message, which can contain text or thinking blocks.

Response example

{
  "id": "msg-example",
  "type": "message",
  "role": "assistant",
  "model": "MODEL_ID",
  "content": [
    {
      "type": "text",
      "text": "OK"
    }
  ],
  "stop_reason": "end_turn",
  "usage": {
    "input_tokens": 12,
    "output_tokens": 1
  }
}

Fields

Field	Type	Description
`id`	string	Message identifier.
`type`	string	Object type `message`.
`role`	string	Message role `assistant`.
`model`	string	Model that processed the request.
`content`	array of object	Generated blocks. Read text or thinking content according to the type.
`content[].type`	string	Block type: `text` for text or `thinking` for reasoning content.
`content[].text`	string	Generated text in a text block.
`content[].thinking`	string	Content of a thinking block.
`content[].signature`	string	Signature returned with a thinking block; preserve it when present.
`stop_reason`	string or null	Stop reason, such as `end_turn` or `max_tokens`.
`usage`	object	Token usage for this request.
`usage.input_tokens`	integer	Input tokens.
`usage.output_tokens`	integer	Output tokens.
`usage.cache_creation_input_tokens`	integer	Cache-write input tokens, when returned.
`usage.cache_read_input_tokens`	integer	Cache-read input tokens, when returned.

Streaming response¶

Set stream to true in the request body to receive generated content incrementally. The service sends server-sent events (SSE) and ends the message with message_stop.

Request example

curl -N -X POST \
  "https://token.moi.matrixorigin.cn/v1/messages" \
  -H "Authorization: Bearer $GENESIS_ACCESS_TOKEN" \
  -H 'Content-Type: application/json' \
  -d '{
  "model": "'"$MODEL_ID"'",
  "max_tokens": 256,
  "messages": [
    {
      "role": "user",
      "content": [
        {
          "type": "text",
          "text": "Reply with OK."
        }
      ]
    }
  ],
  "stream": true
}'

Response example

event: message_start
data: {"type":"message_start","message":{"id":"msg-example","type":"message","role":"assistant","content":[],"usage":{"input_tokens":12,"output_tokens":0}}}

event: content_block_start
data: {"type":"content_block_start","index":0,"content_block":{"type":"text","text":""}}

event: content_block_delta
data: {"type":"content_block_delta","index":0,"delta":{"type":"text_delta","text":"OK"}}

event: content_block_stop
data: {"type":"content_block_stop","index":0}

event: message_delta
data: {"type":"message_delta","delta":{"stop_reason":"end_turn"},"usage":{"output_tokens":1}}

event: message_stop
data: {"type":"message_stop"}

Fields

Field	Type	Description
`type`	string	Event type matching the SSE event name, such as `message_start`, `content_block_delta`, or `message_stop`.
`message`	object	Initial message in `message_start`; see the successful response above for its fields.
`index`	integer	Content block index linking its start, delta, and stop events.
`content_block`	object	Initial block in `content_block_start`; a text block contains its type and initial text.
`content_block.type`	string	Initial block type; `text` for a text block.
`content_block.text`	string	Initial text of the block.
`delta`	object	Content delta or message state update; interpret it by event type.
`delta.type`	string	Content delta type; `text_delta` for text.
`delta.text`	string	Incremental text to append to the same block.
`delta.thinking`	string	Content in a thinking delta, when present.
`delta.stop_reason`	string	Generation stop reason returned by `message_delta`.
`usage`	object	Token usage returned by `message_delta`, when present.
`usage.output_tokens`	integer	Number of output tokens generated.

Buffer and parse complete SSE events. message_start starts the message; process content_block_start, content_block_delta, and content_block_stop by block index. message_delta can include the stop reason and usage, and message_stop ends the message. If the connection closes before message_stop, the content received so far may be incomplete.

Error response¶

A missing model ID returns HTTP 400. Supply the model ID before sending the request again.

Response example

{
  "error": {
    "message": "missing required parameter: model",
    "type": "invalid_request",
    "code": "invalid_request"
  }
}

Fields

Field	Type	Description
`error`	object	Error details.
`error.message`	string	Error description; the example indicates a missing model ID.
`error.type`	string	Error type; may be omitted by a model service.
`error.code`	string or null	Error code; may be omitted by a model service.

Next steps¶

To check the input length before generation, use Count input tokens.

Last updated on Sep 11, 2026