Copilot cannot index vault on Gemini-1.5-flash

[TaiRaven]

Hi there,

Due to my works policies on ChatGPT, I am forced to use Gemini 1.5 Flash as my LLM of choice.

I have connected this, and can chat and send the current note, but clicking the Index vault option causes the indexing to stop at 0/1000 notes. I have run the force re-index vault command, but that has not worked.

This srror pops up, "Error indexing vault: API Key is not provided for the embedding model.

Assuming then that Gemini cannot index the vault?

[logancyang]

Please check your QA setting, you need a working embedding model and your current embedding model is probably set to an OpenAI one. Just use Google's embedding model text-embedding-004, it should work.