feat: add node-llama-cpp library and CLI to GGUF models

[giladgd]

This PR adds a code snippet of using node-llama-cpp on GGUF models, and also CLI commands to:

• Chat with the model
• Estimate the model compatibility with the current hardware without downloading it

node-llama-cpp makes using llama.cpp in Node.js easy, and also provides convenient CLI commands that can be used without even installing it (assuming Node.js is installed).

I couldn't find where to add the icon for node-llama-cpp, so in case it's not in this repo, you can use this icon.

[Vaibhavs10]

Thanks for the PR @giladgd - It's already looking very good.

re: LocalApps can you send the SVG of the logo for node-llama-cpp as well.

More meta comment - I think this would fit more generally for LocalApps atm, but let's wait to see what others think as well: cc: @julien-c

[Vaibhavs10]

Are there any examples of where this is used? I don't see any tagged repos here: https://huggingface.co/models?other=node-llama-cpp

Note: Whichever repo has the tag node-llama-cpp would have the snippet you defined.

[osanseviero]

We should only have node-llama-cpp as a local app option, not a library, for consistency with the rest

[Vaibhavs10]

Yes! read this: #949 (review)

[julien-c]

agree

[giladgd]

You can use node-llama-cpp to run llama.cpp inside a Node.js project in a similar manner that you can use llama-cpp-python to run llama.cpp inside a Python project, so I think it's useful to have a code snippet for this.
I see that llama-cpp-python appears on all llama.cpp compatible GGUF model repos, regardless of whether the repo has a llama-cpp-python tag (for example, this repo).

Since node-llama-cpp also makes it possible to use it as a CLI to interact with models without installing anything,
I think both cases are worth having an easy to use snippet.

Is there a way to make both accessible?

[julien-c]

IMO let's first ship in local apps and we can always revisit later

[giladgd]

Alright. I removed the library code snippet

[osanseviero]

Very nice 🔥

[osanseviero]

We should only have node-llama-cpp as a local app option, not a library, for consistency with the rest

[giladgd]

@Vaibhavs10 You can use this SVG icon in places that need a small icon, and this SVG logo for places that need a medium-sized logo (note that this logo's colors change depending on the color-scheme CSS property used on the image)

[Vaibhavs10]

Thanks a lot for iterating here @giladgd - really appreciate it! We'll have a new release of LocalApps soon. There's an update required from our end (on our internal codebase) - this PR will be merged along with that!

(Internal note: the SVGs are present here: #949 (comment))

cc: @enzostvs @krampstudio for vis 🤗

[giladgd]

@Vaibhavs10 Do you know when it will show up on GGUF models?
I saw that you added Ollama yesterday, but node-llama-cpp is not there.

[julien-c]

@giladgd it's been deployed now!

cc @Vaibhavs10

[giladgd]

@julien-c I don't see it on the drop-down menu on my end (also when I'm not logged in).
I've tested on this model:

It should be available on all models where llama.cpp is available, since it supports all llama.cpp models.

[julien-c]

you need to select it in hf.co/settings/local-apps

[giladgd]

@julien-c Can you show it by default alongside llama.cpp and Ollama?
node-llama-cpp makes it easy to run llama.cpp models locally without installing anything using a single npx command, so I think having it shown by default and also for unauthenticated users would be very helpful.