Based on the pixegami/langchain-rag-tutorial project, langchain-rag-llama_parse adds several features. The main one is the implementation of Llama-Parse, which expands the range of documents accepted for data, previously limited to markdown files. Furthermore, the agent creation process (search databases) has been improved, as has the execution of prompts.

Install dependencies.

pip install -r requirements.txt

Configure Agents.py and create agents

Agents.py

Configure run.py and run the agents

run.py

You'll also need to set up an OpenAI and Llma Parse account (and set the keys on Agents.py and run.py) for this to work.