Pavol Rusnak

> (This is for Linux, don't know about MacOS). Yes, it works for macOS too. I updated my suggestion here to contain the option: https://github.com/ggerganov/llama.cpp/pull/338#discussion_r1143489500

I went ahead and implemented the suggestions from above and rebased/squashed on top of the current master.

Thanks @gjmulder for computing the hashes. Merged!

Please open new pull requests if something is wrong. > Maybe using a naming scheme like "ggml2-model-q4_0.bin" would be good to differentiate between the versions and avoid confusion. We should...

Can you provide some tests that proof the change actually does improve the situation? We already have these 3 mechanisms in master (after https://github.com/ggerganov/llama.cpp/pull/801 merge), so I am wondering if...

This is not really a LNbits bug or issue, it's an issue with the deployment at lnbits.com. Also it seems there is very little will to get this fixed. I...

Let's close this in favor of https://github.com/lnbits/lnbits/issues/1941 which contains more info.

Another option is just using `fprintf(stderr, ...)` everywhere except in that one particular case that prints tokens from the model. According to [POSIX standard](https://unix.stackexchange.com/questions/331611/do-progress-reports-logging-information-belong-on-stderr-or-stdout), stderr is not limited for errors,...

Please fix build failures. If the build failures turn out to be more problematic, we can extract the first commit and submit it as a separate pull request, which can...

I went ahead and extracted the first commit (including my suggestion from above https://github.com/ggerganov/llama.cpp/pull/953#discussion_r1167001496) as Pull Request https://github.com/ggerganov/llama.cpp/pull/970