I have a really strange issue where sometimes, with a large enough system message (or at least that's how I was able to reproduce it), I'm receiving an completely empty response from the completions endpoint (streaming).

The even weirder thing is that there's also no log showing up in the portkey UI, but I do receive a trace ID in the response headers. 46698b1e-479c-4287-a612-c249e20dee22 is such an example.

I may be overlooking it if it exists, but I'd love to have a way see my current logs usage so I can understand when I need to upgrade, or how many logs I've already consumed this month for billing purposes!

Hey friends, love Portkey so far especially since it integrates an LLM gateway and an LLM observability solution into a single product! A few issues I've discovered so far:

• Integrating Sonnet 3.5, I've noticed that even though Portkey says it conforms to the OpenAI spec, for streaming responses I'm receiving an end_turn as finish reason, instead of stop like OpenAI would do
• While the docs say I can attach the user property to the /v1/chat/completions request, the user property was only populated in the Portkey UI when I was using the _user metadata KV
• Not a bug per-se, but more of a usability issue, but it took me forever to figure out why the "Create" button for my new Guardrail is disabled -> I didn't put a name for the Guardrail, but there was no indication in the UI that I was missing the name