Crew Studio is an amazing way to scaffold your Crews or even build some of them (for now the less complex ones) end-to-end with no-code. We use it every day! Crew Studio allows you to chat with our Crew Assistant and describe the problem that you are trying to solve with a Crew. It will generate the agents and tasks, select tools and suggest the inputs based on the provided information. After you generate the flow, you can download the code to connect it to custom APIs and later deploy to the platform!
That being said there are a few things that you need to configure first!
- Go to LLM Connections tab
- Create a new LLM connection, currently Studio supports only OpenAI models, but we are working on expanding it with other models soon!
- Enter “Connection Name”, eg. OpenAI
- Select your model provider: openai or azure
- Select models you’d like to use in your Crews generated by the Studio. We suggest at least gpt-4o, o1-mini and gpt-4o-mini
- In the Environment Variables section add a variable called OPENAI_API_KEY and provide your API key. If you want to configure Azure OpenAI, refer to this article.
- The configuration should look something like on the screenshot below. Click “Add Connection”
- Once you do that you’ll see your connection added to the list
- Now, in the Menu go to Settings → Defaults and configure the LLM Defaults. We recommend the following configuration. Click Save Settings.
You’re good to go! Now you can start using Studio!