How to enable prompt caching with Anthropic model?

[riywo]

Anthropic API requires to specify the explicit parameter called cache_control to enable prompt caching for specific messages: https://docs.anthropic.com/en/docs/build-with-claude/prompt-caching

I'm wondering how to do this with Agno framework.

This also implies that this method needs to updated to calculate the token metrics properly. (I guess putting both cache_*_input_tokens to prompt_tokens_details would be enough.)

agno/libs/agno/agno/models/base.py

Line 1051 in 0ef0d09

def _add_usage_metrics_to_assistant_message(self, assistant_message: Message, response_usage: Any) -> None:

[monali7-d]

Hi @riywo
thanks for reaching out and supporting Agno. I’ve shared this with the team, we’re working through all requests one by one and will get back to you soon.
If it’s urgent, please let us know. We appreciate your patience!