Justine Tunney

Good to know. Thank you.

Looks good to me overall. Is there any chance I could persuade you to vendor the HTTP digest code? What it's doing is simple enough to not merit a dependency....

Is that related to this? https://github.com/ggerganov/llama.cpp/issues/4038

It looks like we need to make our matrix multiplication algorithm more generic on Windows. ``` tinyblasSgemm(m=577, n=577, k=64, lda=64, ldb=64, ldc=577) tinyblasSgemm(m=577, n=577, k=64, lda=64, ldb=64, ldc=577) tinyblasSgemm(m=577, n=577,...

I've just merged PR #153 which makes tinyBLAS image processing w/ LLaVA go 5x faster. Thank @ahgamut for contributing it. NVIDIA's 500mb closed source cuBLAS DLL is still 2x faster...

Thanks for testing that @jbreu!

llamafile is a static binary. If your OS enforces a policy that prevents foreign programs from linking GPU libraries, then there's not a whole lot you can do besides switching...

The `zipalign` command has a pretty good `man` page because I wrote it myself. ``` zipalign(1) General Commands Manual zipalign(1) NAME zipalign – PKZIP for LLMs SYNOPSIS zipalign [FLAG...] ZIP...

I've just made the `--help` flag much more helpful for each program. Every llamafile will now be able to display the rendered man page when that flag is passed in...

Oh last thing worth mentioning. Here's your dev cycle if you edit the `.1` Troff manuals: ``` nano llamafile/llamafile.1 man llamafile/llamafile.1 ``` The way I'm converting them to PDF is:...