Make SuperAgent a ModelProvider

[lightningRalf]

Validations

• [X] I believe this is a way to improve. I'll try to join the Continue Discord for questions
• [X] I'm not able to find an open issue that requests the same enhancement

Problem

I want to be able to use Continue with a SuperAgent / a superagent workflow. https://github.com/continuedev/continue/tree/main/core/llm/llms

Their co-founder / maintainer: homanp will reach out (hopefully)

Solution

Add a new modelprovider: https://github.com/continuedev/continue/tree/main/core/llm/llms Create model providers for their API bases: Workflows: https://docs.superagent.sh/api-reference/api-reference/workflow/invoke Agents: https://docs.superagent.sh/api-reference/api-reference/agent/invoke

Sorry I am not more helpful (no programmer yet)

[homanp]

Co-founder/maintainer of Superagent here. Would be happy to contribute here if needed.

[sestinj]

@homanp @lightningRalf Love the idea, I think this could work very well. I'm new to SuperAgent, so digging through the docs, but my understanding of what could happen here is:

• SuperAgent helps people create and host their own AI assistants
• There are endpoints that can be called which will stream a text response from a given "assistant"
• Each such "assistant" could be used as a "model" in Continue

For both of the endpoints mentioned above, the docs say they return basically empty objects, but it looks like they have a streaming parameter, so will they actually respond with SSE or something similar?