This repository showcases a current bug in the LlamaIndex OpenAIResponses implementation.

When the following conditions are met:

1. A FunctionAgent with tools
2. When OpenAIResponses is the llm
3. When the response is NOT streamed...

LlamaIndex will return an empty response immediately after the tool call instead of fetching the LLM's final response.

To run this example:

1. Copy .env.example to .env
2. Add your OpenAI key to .env
3. Run the script:

$ uv run python main.py

This will result in the following output:

OpenAI response: The result of \(2 \times 3 + 4\) is 10.
OpenAIResponses response: