Anindyadeep
Anindyadeep
Hi, Thank you for the PR, but since there came a major update and the whole pipeline code has been removed with a new module called agents (will release soon)....
Hi, thanks for the suggestion , well a native js library might not be in our roadmap right now for bandwidth issue. But we will be rolling out our apis...
Hi just did an early release of PremSQL. Now you can start the PremSQL server and then use it with js axios or whatever language / lib you want. I...
> maybe add the benchmarks in README? well I am not sure, if I would add the benchmarks in the main readme, I can add it to llama cpp readme...
> Adding schema of all tables in prompt can exceed LLM context limit hence not viable. How PremSQL handles schema with large number of tables? Is there a strategy to...
> Hey, thanks for the question! Unfortunately ~12 seconds per message is about the expected time assuming you are using openAI models. For some broader context: the `add_episode` method does...
Hi, I was profiling each and every ops inside the `add_episode` function and from what I understood, I did some couple of small changes which brought down upsert time of...
> All of these changes will reduce accuracy. `add_episode` is a process that can be done asynchronously outside of an end user's control flow so efficacy takes precedence of latency...
yes great point, even I also have been looking into this project for some time. We will be adding a whole new class of retrievers in the upcoming releases. Thanks...
Yes, I agree