AutoGen Studio for agent teams
AI agent teams that collaborate — by Microsoft
This template does not use a GPU: you can also run it on a dedicated CPU machine if you would rather have more cores and memory than a graphics card.
- Video memory
- não precisa de GPU
- Setup
- ~4 min
- Access
- port 8081
- Billing
- hourly, in BRL
In AutoGen Studio you define roles — researcher, reviewer, executor — and the agents talk to each other until the task is done. It is the most legible way to understand multi-agent orchestration.
AutoGen Studio is Microsoft Research's IDE for building AI agent teams that converse, debate and execute tasks together. Define profiles (coder, reviewer, manager), connect tools (Python, web search) and watch the magic happen.
What it is for
- Split a complex task between agents with distinct roles
- Watch the agent conversation step by step, for debugging
- Wire the agents to your own model, with no per-token cost
- Prototype before writing the real orchestrator
How to deploy
- Create your account and add balance (card or Pix, no subscription).
- In the console, pick the AutoGen Studio template and a machine — the console hides the ones that do not meet the requirement.
- In about 4 minutes the setup finishes and the access address shows up in the panel, on port 8081.
Done? Just destroy the machine and billing stops with it. No contract, no minimum commitment.
FAQ
Is the project still maintained?
It moved into maintenance mode and new development shifted to Microsoft's successor framework. It is still excellent for learning and prototyping; for a new production system, evaluate the successor.
Does it need a GPU?
Only if the model runs on the same machine.
Can I use a local model?
Yes, by pointing it at a vLLM or Ollama with a compatible API.