chore: add Moshi

[Vaibhavs10]

ref: https://huggingface.co/models?other=moshi

[julien-c]

this looks super specific no?

[Vaibhavs10]

Yes! Since there's a mic of safetensors and GGUFs this points to the tokenizer for now - this is the same across all the repos, irrespective of the backend (candle, mlx, PyTorch) - it's the safest way I found without double counting.

[Wauplin]

Any thoughts about having

Suggested change

	countDownloads: `path:"tokenizer-e351c8d8-checkpoint125.safetensors"`,
	countDownloads: `path:"model.safetensors" OR path:"model.q4.safetensors" OR path:"model.q8.safetensors" OR path:"model.gguf" OR path:"model.q4.gguf" OR path:"model.q8.gguf"`,

?

It's long and ugly but at least if a new tokenizer is pushed, it won't break the download count. And it wouldn't duplicate counts if I checked correctly.

Otherwise I'm fine with the current solution, there is no perfect one anyway.

[julien-c]

should we add snippets?

[Vaibhavs10]

should we add snippets?

Yes! I was thinking about it, but wasn't sure how best to support all three backends, since each has a different setup and installation 😅

[julien-c]

but wasn't sure how best to support all three backends

if i'm not mistaken there's a way to specify multiple snippets now, even with titles/collapsible sections etc

[julien-c]

can be in a separate PR though, if you prefer

[julien-c]

pre-approving just in case

[Wauplin]

Yes! I was thinking about it, but wasn't sure how best to support all three backends, since each has a different setup and installation 😅

Is it possible to add mlx or candle tags to the repos themselves and generate snippets based on that?

(agree can be done in a follow-up PR)

[Wauplin]

Thanks for handling it @Vaibhavs10 :)

Approving "as is" (comments above are optional)

[Vaibhavs10]

Cool! ah yeah! Good shout re: MLX + Candle tags, I'll add them tomorrow/ later today and make a follow-up PR 🤗

Thank you for the review and suggestions! ❤️