News for hackers. Stuff that matters.
Re:Llama2.c: inference llama 2 in one file of pure C by pedrovhb on Sunday July 23, 2023 @06:27PM (#36840577)
I've been playing with running some models on the free tier Oracle VM machines with 24GB RAM and Ampere CPU and it works pretty well with llama.cpp. It's actually surprisingly quick; speed doesn't scale too well with the number of threads on CPU, so even the 4 ARM64 cores on that VM, with NEON, run at a similar speed to my 24-core Ryzen 3850X (maybe about half reading speed). It can easily handle Llama 2 13B, and if I recall correctly I did manage to run a 30B model in the past too. Speed for the smaller ones is ~half reading speed or so.

It's a shame the current Llama 2 jumps from 13B to 70B. In the past I tried running larger stuff by making a 32GB swap volume, but it's just impractically slow.

Re:Llama2.c: inference llama 2 in one file of pure C by brucethemoose2 on Sunday July 23, 2023 @07:14PM (#36840955)
Prompt ingestion is too slow on the Oracle VMs.

Also its really tricky to even build llama.cpp with a BLAS library, to make prompt ingestion less slow. The Oracle Linux OpenBLAS build isnt detected ootb, and it doesn't perform well compared to x86 for some reason.

LLVM/GCC have some kind of issue identifying the Ampere ARM architecture (march=native doesn't really work), so maybe this could be improved with the right compiler flags?

Re:Llama2.c: inference llama 2 in one file of pure C by jvickers on Sunday July 23, 2023 @07:48PM (#36841277)
Is it any easier when using Ubuntu on ARM Oracle servers?
Re:Llama2.c: inference llama 2 in one file of pure C by brucethemoose2 on Sunday July 23, 2023 @08:35PM (#36841580)
Nah, I tried Ubuntu too.

The OpenBLAS package was missing on ARM, along with some other dependencies I needed for compilation.

At the end of the day, even with many tweaks and custom compilation flags, the instance was averaging below 1 token/sec as a Kobold Horde host, which is below the threshold to even be allowed as a llm host.

Re:Llama2.c: inference llama 2 in one file of pure C by pedrovhb on Sunday July 23, 2023 @09:04PM (#36841784)
Not sure if that's still the case. I remember having trouble building it a couple of months ago, had to tweak the Makefile because iirc it assumed ARM64 <=> Mac, but I recently re-cloned the repo and started from scratch and it was as simple as `make DLLAMA_BLAS=1`. I don't think I have any special setup other than having installed the apt openblas dev package.
Re:Llama2.c: inference llama 2 in one file of pure C by brucethemoose2 on Sunday July 23, 2023 @10:26PM (#36842413)
IDK. A bunch of basic development packages like git were missing from my Ubuntu image when I tried last week, and I just gave up because it seemed like a big rabbit hole to go down.

I can see the ARM64 versions on the Ubuntu web package list, so... IDK what was going on?

On Oracle Linux, until I changed some env variables and lines in the makefile, the openblas build would "work," but it was actually silently failing and not using OpenBLAS.

Re:Llama2.c: inference llama 2 in one file of pure C by summarity on Monday July 24, 2023 @02:52AM (#36843935)
If you're running on Ampere, using llama.cpp is probably not ideal. While it's optimized for ARM, Ampere has native acceleration for workloads like this: https://cloudmarketplace.oracle.com/marketplace/en_US/adf.ta...