Last Update: 2025/3/26

Gizmo OpenAI Action API

The Gizmo OpenAI Action API does not support streaming and only support > gizmo-gpt4o. It involves calling a third-party API, which may require multiple interface requests triggered by the action click event.

Endpoint

POST https://platform.llmprovider.ai/v1/action

Request Headers

Header	Value
Authorization	Bearer YOUR_API_KEY
Content-Type	application/json

Request Body

The request body should be a JSON object with the following parameters:

Parameter	Type	Description
gizmo_id	string	Unique identifier for the Gizmo instance.
action	string	Specifies the action to be performed.
messages	array	List of messages in the conversation.
parent_message_id	string	(Optional) ID of the parent message in the conversation.
conversation_id	string	(Optional) ID of the conversation.
stream	bool	(Optional) Whether to enable streaming response (false by default).
model	string	The model to use (e.g., gizmo-gpt-4o).
history_and_training_disabled	bool	(Optional) Whether to disable history and training data storage.

---

Message Structure (ChatCompletionMessage)

Parameter	Type	Description
role	string	Role of the message (user, assistant, etc.).
name	string	Name of the entity sending the message.
content	string	Message text content.
attachments	array	(Optional) List of file attachments.
parts	array	(Optional) List of content parts (e.g., images).

---

Attachment Structure (Attachment)

Parameter	Type	Description
id	string	Unique identifier of the attachment.
name	string	Original filename.
size	int64	File size in bytes.
fileTokenSize	int	(Optional) Token size for processing the file.
mimeType	string	MIME type of the file.
width	int	(Optional) Image width (if applicable).
height	int	(Optional) Image height (if applicable).
oss_url	string	(Optional) URL of the file stored in OSS.

---

Part Structure (Part)

Parameter	Type	Description
name	string	Name of the content part.
asset_pointer	string	Reference to the asset location.
size_bytes	int	Size of the part in bytes.
width	int	Image width (if applicable).
height	int	Image height (if applicable).
mimeType	string	MIME type of the part.
image_data	string	(Optional) Base64-encoded image data.
oss_url	string	(Optional) URL of the asset stored in OSS.

Example Request

{
  "messages": [
    {
      "role": "user",
      "content": "1+2+3+....1000=?"
    }
  ],
  "model": "gizmo-gpt-4o",
  "stream": false,
  "gizmo_id": "g-HMNcP6w7d"
}

Response Body

The response body will be a JSON object containing the generated completions and other metadata.

Example Request

Shell

curl -X POST https://platform.llmprovider.ai/v1/action \
  -H "Authorization: Bearer $YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "messages": [
      {
        "role": "user",
        "content": "1+2+3+....1000=?"
      }
    ],
    "model": "gizmo-gpt-4o",
    "stream": false,
    "gizmo_id": "g-HMNcP6w7d"
  }'

nodejs

const axios = require('axios');
const apiKey = 'YOUR_API_KEY';
const url = 'https://platform.llmprovider.ai/v1/action';

const data = {
    stream: false,
    gizmo_id: 'g-HMNcP6w7d',
    model: 'gizmo-gpt-4o',
    messages: [
        {
            role: 'user',
           "content": '1+2+3+....1000=?'
        }
    ]
};

const headers = {
    'Authorization': `Bearer ${apiKey}`,
    'Content-Type': 'application/json'
};

axios.post(url, data, { headers })
.then(response => {
    console.log('Response:', response.data);
})
.catch(error => {
    console.error('Error:', error);
});

python

import requests
import json

api_key = 'YOUR_API_KEY'
url = 'https://platform.llmprovider.ai/v1/action'

headers = {
    'Authorization': f'Bearer {api_key}',
    'Content-Type': 'application/json'
}

data = {
    'stream': False,
    'gizmo_id': 'g-HMNcP6w7d',
    'model': 'gizmo-gpt-4o',
    'messages': [
        {
            'role': 'user',
            "content": '1+2+3+....1000=?'
        }
    ]
}

response = requests.post(url, headers=headers, data=json.dumps(data))

if response.status_code == 200:
    print('Response:', response.json())
else:
    print('Error:', response.status_code, response.text)

For any questions or further assistance, please contact us at [email protected].