Guide of how to build a vision-enabled agent with autogen + llava
With break down of 160-page GPT4V capabilities & visual prompting techniques

https://youtu.be/JgVb8A6OJwM

Microsoft autogen multi agent framework is very interesting;
It enable us to set human feedback input for agents, and lots of control

https://youtu.be/Bq-0ClZttc8?si=mB6zDQOd5BbO-heA

Tutorial of how to build a web agent to control browser, complete complex web tasks & scrape almost anything

https://youtu.be/IXRkmqEYGZA?si=3SgBLWRIe-N2TKfj

https://youtu.be/qKtM2AlDTs8

I was trying to get my custom GPT chatbot respond image/multi media reference, and used llamaindex for creating index & retrieving data 🙂

Hope it helps, DM me for any question pls!

also - how can i change the system prompt for the agent?

Hey team, I was trying to use the codeinterpreter tool, but got error:
ModuleNotFoundError: No module named 'llama_hub.tools.tool_spec'

Can you please help?