Anthropic (claude) support

[omri-suissa-clearmash]

Can we use SharpToken for Anthropic? I could not find if claude is using "cl100k_base" or other encoding

[dmitry-brazhenko]

Hello @omri-suissa-clearmash !

Could you share a bit more what is claude or Anthropic? What is there and how it works?

Thanks

[omri-suissa-clearmash]

@dmitry-brazhenko
claude is the LLM of Anthropic (https://www.anthropic.com/).
This is what I could find:
https://github.com/anthropics/anthropic-sdk-python/blob/e84645b07ca5267066700a104b4d8d6a8da1383d/src/anthropic/_tokenizers.py

[dmitry-brazhenko]

Thanks for sharing.

I will check that. Probably they use some already existing encoding (cl100k_base) or just some custom one. I will check.

[omri-suissa-clearmash]

@dmitry-brazhenko also found this: https://github.com/19h/claude_tokenizer (rust)

[dmitry-brazhenko]

Hello @omri-suissa-clearmash !

I checked the algorithm. Seems that there is a difference, but it can be potentially implemented into Sharptoken lib.
I will try to do that within a few days

[Mrgaton]

Any updates on this?