LocalLLaMA

11 readers

4 users here now

Community to discuss about Llama, the family of large language models created by Meta AI.

founded 2 years ago

MODERATORS

communick@poweruser.forum

Mozilla Llamafile - bundling one model & llama.cpp into a single executable (alien.top)

submitted 2 years ago by Danmoreng@alien.top to c/localllama@poweruser.forum

4 comments fedilink hide all child comments

Just read about this project on Twitter and it sounds really interesting:

https://github.com/mozilla-Ocho/llamafile

What do guys think, might be even more simple than Ollama?

you are viewing a single comment's thread
view the rest of the comments

[–] mcbagz@alien.top 1 points 2 years ago

It is probably better to ask here than to create a new post:

They have noted that some Windows computers cannot run the programs that are larger than 4gb. Can anyone put together a really straightforward explanation of downloading and running llamafile not bundled with weights and connecting it to a set of weights?