Repository navigation
Replies: 1 comment
|
"Failed to fetch" is the browser's own error. It means the create request got no HTTP response at all; an HTTP error from LocalAI would show its message instead. So first check the frontend's log (with For RAG in distributed mode with several frontends, you need v4.12.0 or later. Before that, each frontend kept its own in-memory list of collections, so a collection created on one frontend gave LOCALAI_AGENT_POOL_VECTOR_ENGINE=postgres
LOCALAI_AGENT_POOL_DATABASE_URL=postgresql://user:pass@postgres:5432/localrecall?sslmode=disable
On current master these variables are renamed to |
Uh oh!
There was an error while loading. Please reload this page.
We have a frontend, and two workers with
llama.cppbackends in Distribution mode (PostgreSQL + NATS).How to set up RAG? We can't even create a Knowledge base collection:
All reactions