Chat completion

Chat completion by Reigent.

• URL: /v1/chat/completions

• Method: POST

• Headers:

  Key	Value
  Authorization	Bearer rei-agent-secret-token

• Request:

  ---

  model string Optional

  Default: Agent's Configured Model

  Allowed values:

  • openai/gpt-4.1-mini
  • gpt-image-1
  • x-ai/grok-3-beta
  • google/gemini-2.5-flash

  ---

  messages array (min length: 1) Required

  Show message structure

  Each message object contains:

  role string Required
  The role of the message author.
  Allowed values: "system", "user", "assistant", "tool"

  content string or array Required
  The contents of the message. Can be:

  • Simple text string
  • Array of content parts (for multimodal inputs)

  Show content parts structure

  Each content part object contains:

  type string Required
  The type of content part.
  Allowed values: "text", "image_url", "file"

  text string Conditional
  Text content (required when type is "text")

  image_url object Conditional
  Image URL details (required when type is "image_url")
  Contains:

  • url string Required
    The URL of the image

  
  

  file object Conditional
  File details (required when type is "file")
  Contains:

  • filename string Required
    The name of the file

  • file_data string Required
    The buffer content of the file

  name string Optional
  An optional name for the participant

  tool_call_id string Optional
  Required when role is "tool"

  tool_calls array Optional
  Tool calls made by the assistant

  ---

  temperature number Optional

  Range: 0 to 2 Default: 1

  ---

  max_tokens integer Optional

  Minimum: 1

  ---

  top_p number Optional

  Range: 0 to 1

  ---

  n integer Optional

  Minimum: 1

  ---

  seed number Optional

  Range: 0 to 2^53-1

  ---

  stream boolean Optional

  ---

  stop string or array[string] Optional

  ---

  presence_penalty number Optional

  Range: -2.0 to 2.0

  ---

  frequency_penalty number Optional

  Range: -2.0 to 2.0

  ---

  logit_bias object Optional

  Key format: Token IDs as string numbers Value range: -100 to 100

  ---

  logprobs boolean Optional

  ---

  top_logprobs number Optional

  ---

  user string Optional

  ---

  response_format object Optional

  • type string
    • Allowed values: "text", "json_object"

  ---

  tools array Optional
  A list of tools the model may call

  Show tool structure

  Each tool object contains:

  type string Required
  Must be: "function"

  function object Required
  The function definition

  function.name string Required
  The name of the function

  function.parameters object Required
  The parameters the function accepts

  ---

  tool_choice string or object Optional
  Controls which tool is called
  Allowed string values: "none", "auto"
  Or specify a tool with:

  {
    "type": "function",
    "function": {
      "name": "tool_name"
    }
  }

---

• Sample Request

1. Type: Text

Show sample

{
    "messages": [
        {
            "role": "user",
            "content": "Hello, can you help me with my research?"
        }
    ],
    "tools": [
        {
            "type": "function",
            "function": {
                "name": "get_weather",
                "description": "Get current temperature of given location",
                "parameters": {
                    "type": "object",
                    "properties": {
                        "location": {
                            "type": "string",
                            "description": "City and country (e.g. Paris, France)"
                        }
                    },
                    "required": ["location"],
                    "additionalProperties": false
                },
                "strict": true
            }
        }
    ]
};

2. Type: Image (in URL)

Show sample

{
    "messages": [
        {
            "role": "user",
            "content": [
              {
                "type": "text",
                "text": "Hello, can you help me with my research?"
              },
              {
                "type": "image_url",
                "image_url": {
                  "url": "https://test.png"
                }
              },
            ]
        }
    ],
    "tools": []
};

3. Type: Image (in Base64)

Show sample

{
    "messages": [
        {
            "role": "user",
            "content": [
              {
                "type": "text",
                "text": "Hello, can you help me with my research?"
              },
              {
                "type": "image_url",
                "image_url": {
                  "url": "data:image/png;base64,iV..."
                }
              },
            ]
        }
    ],
    "tools": []
};

4. Type: Docs (PDF)

Show sample

{
    "messages": [
        {
            "role": "user",
            "content": [
              {
                "type": "text",
                "text": "Hello, what's inside the PDF?"
              },
              {
                "type": "file",
                "file": {
                  "filename": "Sample File Name",
                  "file_data": "data:application/pdf;base64,JVBERi0xLjMNCiXi48/....",
                }
              }
            ]
        }
    ],
    "tools": []
};

---

• Response:

1. Without Tools

Show sample

{
  "choices": [
    {
      "index": 0,
      "message": {
        "content": "Hello! How can I assist you today?",
        "role": "assistant"
      }
    }
  ]
}

2. With Tools

Show sample

{
  "choices": [
    {
      "index": 0,
      "message": {
        "content": "",
        "role": "assistant",
        "tool_calls": [
          {
            "id": "call_zSIBPi4QKxjkpAewfi5YbTnI",
            "type": "function",
            "function": {
              "name": "get_weather",
              "arguments": "{\"location\":\"Paris, France\"}"
            }
          }
        ]
      }
    }
  ]
}

---

• Error

Response Code	Reason
400	Validation Error
401	Unauthorized
404	Agent not found