Open jbilcke opened 1 year ago
It would be nice to be able to predict (eg. by taking the worst-case scenario) the usage and token consumption of the LLM vendor.
Some random ideas:
It would be nice to be able to predict (eg. by taking the worst-case scenario) the usage and token consumption of the LLM vendor.
Some random ideas: