Skip to main content
POST
/
v1
/
research-agent
Research Agent
curl --request POST \
  --url https://agents.bigdata.com/v1/research-agent \
  --header 'Content-Type: application/json' \
  --header 'X-API-Key: <api-key>' \
  --data '
{
  "message": "Type your research question here. Example: Summarize recent developments in quantum computing research.",
  "research_effort": "lite",
  "model_name": "base",
  "persistence_mode": "disabled"
}
'
{
  "chat_id": "<string>",
  "message": {
    "content": "<string>",
    "message_id": "<string>",
    "role": "assistant",
    "type": "THINKING"
  }
}

Documentation Index

Fetch the complete documentation index at: https://docs.bigdata.com/llms.txt

Use this file to discover all available pages before exploring further.

Authorizations

​
X-API-Key
string
header
required

API key for authentication.

Body

application/json

Request to execute a research query.

​
message
string
required

Your research question or request.

Minimum string length: 1
​
research_effort
enum<string>
required

Controls research depth. Use 'lite' for quick answers or 'standard' for thorough analysis.

Available options:
lite,
standard
​
chat_id
string | null

Provide a previous chat ID to continue a conversation.

​
from_checkpoint_id
string | null

Resume or edit a conversation from a specific checkpoint.

​
model_name
enum<string>
default:base

Selects the model capability tier. Use "base" for the default routing strategy. Use "pro" to allow the workflow to use the most capable available model when higher reasoning or accuracy is required. LLM Model selection is dynamic and may vary based on task type, availability, failover, and internal performance benchmarks.

Available options:
base,
pro
​
persistence_mode
enum<string>
default:disabled

Set to 'enabled' to save conversation history for follow-up questions.

Available options:
enabled,
disabled
​
structured_output_schema
Structured Output Schema · object

JSON Schema for structured data extraction from the research results.

​
tools_configs
Tools Configs · object

Configuration for the search tool, including filters and ranking.

Response

ResearchAgentResponse · object | null

Streaming SSE response with research results.

A single streaming event from the Research Agent.

​
chat_id
string
required

Identifier for this conversation.

​
message
ThinkingMessage · object
required

The agent's intermediate reasoning while researching.