Installation and deployment:
-
Copy .env.example to .env and fill it with your apikey and you frontend uri.
-
Execute those commands:
-
pip install -r requirements.txt -
PYTHONPATH=. uvicorn main:app -
uvicorn main:app
- get session_id on
GET /chat/init
- connect to websocket on
WS /chat/socket/{session_id} - streaming responses:
POST /chat/stream/{session_id}
- retreive recall memory stats :
GET /memory/{session_id}/recall/stats - TO DO :
- archival memory
- search on memory
Using docker:
docker build -t memgpt .docker run -v $PWD:/usr/src/app -p 8000:8000 --name memgpt0 memgpt