Eric Xanderson

just got dual A6000 from last gen card with 96GB total. Would love to max that out. Can probably help with testing.

[marmolite](https://en.wiktionary.org/wiki/marmolite) might combine the

@luckyrobot Thanks. This appears to work but it seems like such a monkey patch for this library.

looks like @apollo-client @ 3.6.2 installs ts-variant @ 0.10.2 @benjamn

have same issue, but with ```$apollo.mutate```, I can confirm options ```{ errorPolicy: 'all' }``` is passed into the ```client.mutate``` and i've tried to set it as the ```defaultOptions```, function but...

this is a great software for Windows. Sad to hear no future development. I'm definitely missing the ability to arbitrarily put windows into splits. Great job though.

> [@disarticulate](https://github.com/disarticulate) You can use Windows Terminal, mate. It supports splitting windows. Although I have switched to it, I really miss Conemu :( Not the features that are most interesting....

great timing on this feature. is the docker images lobehub/lobe-chat-database updated?

was able to test this out. After figuring out a correct embedding model for ollama: xiaoming/bce-embedding-base_v1 and trying both the proxy OpenAI Url and the ollama url: ``` DEFAULT_FILES_CONFIG=embedding_model=ollama/xiaoming/bce-embedding-base_v1 DEFAULT_FILES_CONFIG=embedding_model=openai/xiaoming/bce-embedding-base_v1...

(I've never developed a VScode extension) There's probably two options: 1. Forward the necessary ports for ollama. Not sure if this is a two way street, but I know VSCode...