That's a good tbh. I wonder whether we should implement instrumentations for LLMs "hosting solutions" or for specific LLMs (E.g. LLaMa/Falcon) and ignore the hosting solution (not sure if that's even possible though as it sort of dictates the inference api).
We definitely have pgvector on our roadmap (which tbh I think we better publish in the repo). For Graphana tempo, it's just a matter of making sure that it works as a destination - we'll do it today/tomorrow.
I totally get you; we're struggling to find the correct wording. As an engineer, I also bump into the "gradual (temporal) rollout" as the first association of "Feature Flags".
What wording would make sense to you now that you understand the vision?
It goes beyond a feature toggle. Pricing Tiers/Packages tend to consist of more "sophisticated" pricing mechanisms, usually split into Licensed Features (simple toggles like SAML enabled and limits like Seat Count) and Metered Features (either quotes/limits or pay as you go). We're currently working on the metering functionality together with billing systems integration. That's the difference, IMO, from the common Feature Flags solution.
Thanks for the comments; they are much appreciated and help us shape our messaging and understand what's not going through.
That's a good tbh. I wonder whether we should implement instrumentations for LLMs "hosting solutions" or for specific LLMs (E.g. LLaMa/Falcon) and ignore the hosting solution (not sure if that's even possible though as it sort of dictates the inference api).
wdyt?