Skip to main content
Quest provides a REST API through Flask that allows you to interact with the RAG engine programmatically. All endpoints accept and return JSON data.

​
Base URL

When running locally: 
http://localhost:5000

​
Endpoints

​
GET /

Renders the main web interface. Response
  • Returns HTML page (index.html template)

​
POST /search

Submit a query to the RAG engine and get a solution.
​
query
string
required
The coding problem or question to search for
​
mode
string
default:"general"
Inference mode: general or reasoning
Response
​
response
string
The generated solution or explanation
Error Response
​
error
string
Error message if the request fails
Example Request
curl
curl -X POST http://localhost:5000/search \
  -H "Content-Type: application/json" \
  -d '{
    "query": "Two Sum problem",
    "mode": "general"
  }'
Python
import requests

response = requests.post(
    "http://localhost:5000/search",
    json={
        "query": "Two Sum problem",
        "mode": "general"
    }
)

data = response.json()
print(data["response"])
Example Response 
{
  "response": "Generated Solution:\n[Solution content here]"
}

​
POST /stop

Stop the currently running generation process. Response
​
message
string
Confirmation message
Example Request
curl
curl -X POST http://localhost:5000/stop
Python
import requests

response = requests.post("http://localhost:5000/stop")
print(response.json())
Example Response 
{
  "message": "Streaming stopped"
}

​
POST /clear_history

Clear the conversation history stored in the memory buffer. Response
​
message
string
Confirmation message
Example Request
curl
curl -X POST http://localhost:5000/clear_history
Python
import requests

response = requests.post("http://localhost:5000/clear_history")
print(response.json())
Example Response 
{
  "message": "Conversation history cleared"
}

​
GET /get_history

Retrieve the current conversation history. Response
​
history
string
Formatted conversation history with previous queries and responses
Example Request
curl
curl http://localhost:5000/get_history
Python
import requests

response = requests.get("http://localhost:5000/get_history")
print(response.json())
Example Response 
{
  "history": "User: What is dynamic programming?\nSystem: Dynamic programming is...\n"
}

​
POST /set_mode

Change the inference mode between general and reasoning.
​
mode
string
required
Must be either general or reasoning
Response
​
message
string
Confirmation message with the new mode
Error Response
​
error
string
Error message if mode is invalid
Example Request
curl
curl -X POST http://localhost:5000/set_mode \
  -H "Content-Type: application/json" \
  -d '{"mode": "reasoning"}'
Python
import requests

response = requests.post(
    "http://localhost:5000/set_mode",
    json={"mode": "reasoning"}
)
print(response.json())
Example Response 
{
  "message": "Mode set to: reasoning"
}
Error Response 
{
  "error": "Mode must be 'general' or 'reasoning'."
}

​
Error Handling

All endpoints may return HTTP error codes:
  • 400 Bad Request - Invalid parameters or missing required fields
  • 500 Internal Server Error - Server-side error during processing
Error responses include a JSON object with an error field containing a descriptive message.

​
Rate Limiting

There is currently no rate limiting implemented. Queries are processed sequentially by the RAG engine.
The /search endpoint response time varies based on the inference mode and query complexity. General mode typically responds in 15-20 seconds, while reasoning mode may take 3-4 minutes.

Build docs developers (and LLMs) love

Get started for freeTalk to us