How exactly Llama is accessed?

Hi,
I would like to download Llama-2-70b-chat-ht with the following command but as you can see it gives an error regarding Meta’s license.

$ git clone https://mahmoodn:TOKEN@huggingface.co/meta-llama/Lla
ma-2-70b-chat-hf ${CHECKPOINT_PATH}
Cloning into 'Llama-2-70b-chat-hf'...
remote: Access to model meta-llama/Llama-2-70b-chat-hf is restricted and you are not in the authorized list. Visit https://huggingface.co/meta-llama/Llama-2-70b-chat-hf to ask for access.
fatal: unable to access 'https://huggingface.co/meta-llama/Llama-2-70b-chat-hf/': The requested URL returned error: 403

When I visit Meta’s page, it gives some instructions to download the models, however, this specific version doesn’t exists:

$ llama model download --source meta --model-id Llama-2-70b-chat-ht
usage: llama model download [-h] [--source {meta,huggingface}] [--model-id MODEL_ID] [--hf-token HF_TOKEN] [--meta-url META_URL]
                            [--ignore-patterns IGNORE_PATTERNS] [--manifest-file MANIFEST_FILE]
llama model download: error: Model Llama-2-70b-chat-ht not found

I got confused on how to download that model. How can I request an access to Llama-2-70b-chat-ht?

That model is a gated model, so you cannot download or use it unless you first fill out the necessary information on this page and get approval before using the token.
In your case, it seems that you have already used the token, so once you have been authorized, you should be able to use the first method after that.

This topic was automatically closed 12 hours after the last reply. New replies are no longer allowed.