Live data from Hacker News

llama.cpp

llama.app

181–182 of 182 posts

Re: llama.cpp

#181

Earlier quoted context omitted.

> If you think Ollama is better, use it! It's not that simple. An army of dweebs from HN will harass you if they find out you're using Ollama. It's getting to the point you can't use it at work. They all link the same article. I never care about it. Forking software was one of the original ideas with GitHub - it's called Ollama . Anyone curious enough to deep dive can find out about the name, and how Ollama came to b…

Why are you posting from multiple accounts?

The multiple accounts, the fanatcism. It’s… weird.

Re: llama.cpp

#182
post #24

llama.cpp works pretty well for me on the Framework 13 laptop, but the current era of "move fast, break things, rarely fix" (sorry, that's how it feels), bites here quite a bit. Two examples: - https://github.com/ggml-org/llama.cpp/pull/25863 Someone's few lines change broke the native (ROCm) support for the AMD GPU inside Framework (and other integrated systems), and any rollback or proper fix is pending for almost…

I have a framework 13, but I couldn't imagine running a local llm on it, how do you do it? Do you have a eGPU?

If you have AMD, you can benefit from the unified memory.
Post reply on HN