4 ms·
They gated it to US providers only, fucking garbage
by ahriad 2mo ago
They gated it to US providers only, fucking garbage
- kentonv 2mo agoNot intentionally, we just haven't gotten around to expanding the provider list beyond what we use regularly. There is ollama (local LLM) support in there. But we'll probably expand it to everything pi-agent-core supports, just need some UI work.
- ahriad 2mo agoI'm talking about setting up one provider for the entire OS not about BYOK per user. Right now only cloudflare AI gateway is supported and in it you hardcoded it to US providers. Plus your AI workers and from your model you only supporting two.
- kentonv 2mo agoThe code currently supports Anthropic, OpenAI, Gemini, Workers AI, and ollama as AI providers. It additionally optionally supports sending all requests through Cloudflare AI Gateway as a proxy. The same set of providers is supported either way, but when you configure AI Gateway, we assume you intend to install side-wide API keys into the gateway. The idea here is that you're setting up the gateway for your entire company, and you want all users to use the same keys. So we no longer let individual users configure their own keys in this mode -- instead they can choose from among the providers you have configured site-wide. We will expand the provider list soon.
- indigodaddy 2mo agoI can't get an AI Gateway custom provider to work within CFOS (deployed to my CF account). No matter what custom model path I use in the cfos admin add provider mechanism, I get a model not found error when attempting to use the model in a conversation...
- kentonv 2mo agoYeah sorry, only the providers I listed are supported right now. We'll be widening support soon.
- indigodaddy 2mo agoHow does the cloudflare-based deployment of cfos get updated? If I "redeploy" will it get updates from cfos upstream? Eg, for when you add new stuff like this upstream..
- kentonv 2mo agoWe're adding an update function. Didn't need it yet on day one. :)
- indigodaddy 2mo agoAwesome!