Pamela Fox

I looked at the Bicep validation error but I think that is an unrelated issue, likely due to a newer version of Bicep that's stricter.

I'm unable to replicate this, here's what I see in Edge:  Is this with user login enabled or other features?

Yes please, thank you @jeannotdamoiseaux

Are you trying to get the final token usage used from the whole flow? We display that in the Thought Process tab, and we get that from the "usage" details...

Ah I see, okay. We currently only show the amount of tokens that we embed in a batch, with this output from the prepdocs script: Computed embeddings in batch. Batch...

The current cost estimate section is worth a reference if you haven't started with that already: https://github.com/Azure-Samples/azure-search-openai-demo?tab=readme-ov-file#cost-estimation For the non-multimodal deployment, I believe the primary per-token cost is Azure OpenAI...

As for measuring token count once they're in the index, you could write a script that iterates through the document chunks in the index, then uses tiktoken to compute the...

Thanks @huntingcodes-001 for the PR. The requirements.txt file is generated by this command: `pip-compile --output-file=requirements.txt pyproject.toml` Will that retain the marker for the windows OS, or will it override when...

The other option: requirements.in could bring in just uvicorn, not uvicorn[standard], and then bring in uvloop with the environment marker you have in this PR. Here's what standard brings in:...

Okay, so if we bring in just uvicorn, not uvicorn[standard], plus uvloop with a marker, we also need to decide whether to bring in the other packages from [standard]: *...