News for hackers. Stuff that matters.

A comment on RamaLama.

Re:RamaLama by esafak on Friday January 31, 2025 @09:38AM (#42888104)
Is this useful? Can someone help me see the value add here?
[ Parent | on HN ]
Re:RamaLama by BubbleRings on Friday January 31, 2025 @09:45AM (#42888173)
Well, if you aren’t that great with Docker but you want to try out a variety of LLMs under Docker, how much would this help you? How much trouble is it to enable an LLM to reach outside of a container to make use of your GPU? How much does this tool help with that?
[ Parent | on HN ]
Re:RamaLama by philipswood on Friday January 31, 2025 @09:43AM (#42888143)
I think the post arose from:

https://news.ycombinator.com/item?id=42886768

[ Parent | on HN ]
Re:RamaLama by maxamillion on Friday January 31, 2025 @09:56AM (#42888285)
ramalama can just pull (almost) any arbitrary model off huggingface and run it ... you're not limited to just what ollama has repackaged into their non-standard format
[ Parent | on HN ]
Re:RamaLama by quantumwoke on Friday January 31, 2025 @09:41AM (#42888124)
It was mentioned in the other thread on the front page [1]

[1] https://news.ycombinator.com/item?id=42886680

[ Parent | on HN ]