Skip to content

How can I use downloaded token_encoding models? #36

@Salieri0515

Description

@Salieri0515

Hi
Currently I can't connect to openaipublic.blob.core.windows.net on my server, which causes following errors when setting token_encoding="o200k_base":

requests.exceptions.ConnectionError: HTTPSConnectionPool(host='openaipublic.blob.core.windows.net', port=443): Max retries exceeded with url: /encodings/o200k_base.tiktoken (Caused by NameResolutionError("<urllib3.connection.HTTPSConnection object at 0x7ff371f9b7d0>: Failed to resolve 'openaipublic.blob.core.windows.net' ([Errno -3] Temporary failure in name resolution)"))

I can only download token embedding models manually and copy it to my server. (llm model is fine with locally deployed url and key).
So, what should I do? I tried token_encoding="/path/to/o200k_base" but it didn't work.

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions