kha84

Ah I see. Sorry my bad. Thanks for clarification

@NoVeTe36 yeah, something similar is what I did as well - created a new "data source" on a machine with internet and then moved the installation to a secured server....

I stumbled across the same issue with `VARIADIC` in https://github.com/open-metadata/OpenMetadata. It uses sqlfluff under the hood to build lineage and fails to parse SQLs like: ``` WITH t AS (...

I can back the above issues. I do have a lighthouse configured on Linux VPS (Debian 11, Nebula version is 1.6.1 - installed from standard debian repo), with public IPv4...

I managed to find steps to reproduce this issue: 1) to have my phone in the same home WiFi network as my Linux machine and have Nebula restarted on both....

I have another installation with older version 0.3.11 and there it takes just few tens of Mb: ``` user@7940hs:~$ ollama --version ollama version is 0.3.11 ``` 

Duplicate https://github.com/antimatter15/alpaca.cpp/issues/89

> How so ? @cboursnell seems to be saying he's getting emoji output all the time, the issue you refer does not seem to mention that ? I just looked...

Sorry for the delayed response. Yes, this happens without using Data Explorer feature. Just regular browsing tables and editing records. Unfortunately I cannot say right now what exact user action...

@derbl4ck based on your experience, [Diarizen](https://github.com/BUTSpeechFIT/DiariZen?tab=readme-ov-file) overperforms pyannote (which is currently baked-in into whisperX) in terms of quality?