ContextQMD
Back to Mindsdb

Create Run

docs/minds/rest_api/create_run.mdx

26.1.05.1 KB
Original Source

This API endpoint creates a Run on a specific Thread using the POST method.

Body

<ParamField body='assistant_id' type='string' required>

ID of the Assistant (Mind) that will be used to run the Thread.

</ParamField> <ParamField body='stream' type='array'>

Stream Mind's thoughts by setting this parameter to true.

</ParamField>

Response

<ResponseField name="id" type="string" required> Unique identifier for the run. </ResponseField> <ResponseField name="assistant_id" type="string" required> ID of the Assistant associated with the run. </ResponseField> <ResponseField name="cancelled_at" type="number"> Timestamp when the run was cancelled. </ResponseField> <ResponseField name="completed_at" type="number"> Timestamp when the run was completed. </ResponseField> <ResponseField name="created_at" type="number" required> Timestamp when the run was created. </ResponseField> <ResponseField name="expires_at" type="number"> Expiration time for the run. </ResponseField> <ResponseField name="failed_at" type="number"> Timestamp when the run failed. </ResponseField> <ResponseField name="incomplete_details" type="object"> Details about why the run was incomplete, if applicable. </ResponseField> <ResponseField name="instructions" type="string"> Instructions sent to the Assistant. </ResponseField> <ResponseField name="last_error" type="object"> Details about the last error, if any. </ResponseField> <ResponseField name="max_completion_tokens" type="number"> Maximum number of tokens for the completion. </ResponseField> <ResponseField name="max_prompt_tokens" type="number"> Maximum number of tokens for the prompt. </ResponseField> <ResponseField name="metadata" type="object"> Additional metadata for the run. </ResponseField> <ResponseField name="model" type="string"> Model name used for the run. </ResponseField> <ResponseField name="object" type="string" required> Type of the returned object (e.g., "thread.run"). </ResponseField> <ResponseField name="required_action" type="object"> Actions required before proceeding. </ResponseField> <ResponseField name="response_format" type="string"> Format of the response. </ResponseField> <ResponseField name="started_at" type="number"> Timestamp when the run started. </ResponseField> <ResponseField name="status" type="string" required> Current status of the run (e.g., "queued", "in_progress"). </ResponseField> <ResponseField name="thread_id" type="string" required> ID of the thread associated with the run. </ResponseField> <ResponseField name="tool_choice" type="string"> Chosen tool for the run. </ResponseField> <ResponseField name="tools" type="array"> List of tools used during the run. </ResponseField> <ResponseField name="truncation_strategy" type="string"> Strategy used to truncate input. </ResponseField> <ResponseField name="usage" type="object"> Usage statistics for the run. </ResponseField> <ResponseField name="temperature" type="number"> Sampling temperature for randomness. </ResponseField> <ResponseField name="top_p" type="number"> Top-p sampling parameter. </ResponseField>

Authorization

A valid API key must be passed in the Authorization header:

Authorization: Bearer MINDS_API_KEY

Generate your API key here.

Path Parameters

<ParamField body='thread_id' type='string' required>

ID of the Thread to run.

</ParamField> <RequestExample>
shell
curl -X POST "https://mdb.ai/threads/mdb_thread_4d3f3805c95f4c1a9d77d32cbdfbe3d9/runs" \
  -H "Authorization: Bearer MINDS_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "assistant_id": "<name of the mind that you created>",
    "stream": true
  }'
python
from openai import OpenAI
client = OpenAI(
    api_key='MINDS_API_KEY',
    base_url='https://mdb.ai/'
)

run = client.beta.threads.runs.create(
  thread_id="thread_abc123",
  assistant_id="<name of the mind that you created>"
)

print(run)
js
import OpenAI from "openai";

const openai = new OpenAI(
    api_key='MINDS_API_KEY',
    baseURL='https://mdb.ai/'
);

async function main() {
  const run = await openai.beta.threads.runs.create(
    "thread_abc123",
    { assistant_id: "<name of the mind that you created>" }
  );

  console.log(run);
}

main();
</RequestExample> <ResponseExample>
json
{
    "id": "mdb_run_8b425f3bf0a04ca584ad11106d37330c",
    "assistant_id": "mdb_asst_cfde04f34a5e47548b705fbf091e06bc",
    "cancelled_at": null,
    "completed_at": null,
    "created_at": 1718885800,
    "expires_at": null,
    "failed_at": null,
    "incomplete_details": null,
    "instructions": "",
    "last_error": null,
    "max_completion_tokens": null,
    "max_prompt_tokens": null,
    "metadata": null,
    "model": "",
    "object": "thread.run",
    "required_action": null,
    "response_format": null,
    "started_at": null,
    "status": "queued",
    "thread_id": "mdb_thread_cfa0fac64cd94faab3ae269d768c0cb0",
    "tool_choice": null,
    "tools": [],
    "truncation_strategy": null,
    "usage": null,
    "temperature": null,
    "top_p": null
}
</ResponseExample>