The Discord link is broken, in that it links to the server directly rather than to an invitation to join the server, which prevents new members from joining.
I’m glad to see llamafile being resurrected. A few things I hope for:
1. Curate a continuously extended inventory of prebuilt llamafiles for models as they are released
2. Create both flexible builds (with dynamic backend loading for cpu and cuda) and slim minimalist builds
3. Upstreaming as much as they can into llama.cpp and partner with the project
Crazier ideas would be:
- extend the concept to also have some sort of “agent mode” where the llamafiles can launch with their own minimal file system or isolated context
- detailed profiling of main supported models to ensure deterministic outputs
Ollama is a model manager and pretty interface for llama.cpp, llamafile is a cross-platform packaging tool to distribute and run individual models also based on llama.cpp
This is great news. Given the proliferation of solid local models, it would be cool if llamafile had a way to build your own custom versions with the model of your choice.
The Discord link is broken, in that it links to the server directly rather than to an invitation to join the server, which prevents new members from joining.
Tips:
And:I’m glad to see llamafile being resurrected. A few things I hope for:
1. Curate a continuously extended inventory of prebuilt llamafiles for models as they are released 2. Create both flexible builds (with dynamic backend loading for cpu and cuda) and slim minimalist builds 3. Upstreaming as much as they can into llama.cpp and partner with the project
Crazier ideas would be: - extend the concept to also have some sort of “agent mode” where the llamafiles can launch with their own minimal file system or isolated context - detailed profiling of main supported models to ensure deterministic outputs
Love the idea!
justine tunney gave a great intro to Llamafile at AIE last year if it helps anyone: https://www.youtube.com/watch?v=-mRi-B3t6fA
Really exciting to see Mozilla AI starting up and I can't wait to see where the next generation takes the project!
People are so uninformed, they don't know you are Mozilla AI's star employee.
s/are/was
I don't know if you were informed but you realize that jart is no longer at Mozilla anymore and now at Google Inc?
how is this different from ollama? for me the more/open the merrier.
Ollama is a model manager and pretty interface for llama.cpp, llamafile is a cross-platform packaging tool to distribute and run individual models also based on llama.cpp
This is great news. Given the proliferation of solid local models, it would be cool if llamafile had a way to build your own custom versions with the model of your choice.
great stuff, working on something around agentic tooling and hope to collab with Mozilla AI as well in the future as they share the same values I have
go get that investor money i guess?