## Share the output returned by the last node
Q: who won the 2024 MLB world series?
A: The 2024 MLB World Series hasn't happened yet! The season is still ongoing. I can't predict the future, but I can definitely help you stay updated on the exciting MLB season!
## Information on your n8n setup
- n8n version: 1.65.2
- Database: Postgres
- n8n EXECUTIONS_PROCESS: (not seeing this setting anywhere)
- Running n8n via Docker on Coolify
- Operating system: Ubuntu 22.04
How did you configure your AI Agent? How are you instructing it to use the tools?
Can you share your workflow so we can try to reproduce this?
Thank You
Tip for sharing your workflow in the forum
Pasting your n8n workflow
Ensure to copy your n8n workflow and paste it in the code block, that is in between the pairs of triple backticks, which also could be achieved by clicking </> (preformatted text) in the editor and pasting in your workflow.
```
<your workflow>
```
Make sure that you’ve removed any sensitive information from your workflow and include dummy data or pinned data as much as you can!
Same question / issue over here. How do we get the AI Agent node tool calling to work with Gemini? It seems that function calling is supported by Gemini (and Vertex).
I confirm, Gemini flash does not run the tool in my case, even with the explicit mention of the name of the tool and the instructions to run it and all the send delails.
ChatGPT does the job, but sometimes misses tool launches, so I abandoned it.
I solved this problem with Gemini Flash by generating JSON in otput response with a lot of complex parsing. but i got near 100% results in tests.