lavague-ai / LaVague

Large Action Model framework to develop AI Web Agents
https://docs.lavague.ai/en/latest/
Apache License 2.0
5.5k stars 506 forks source link

Add more detailed token consumption output to `lavague-tests` #487

Closed paulpalmieri closed 4 months ago

paulpalmieri commented 4 months ago

In order to do more detailed benchmarks and comparison, having more details than just the total number of consumed tokens would be helpful.

Ideally a breakdown by input/output tokens with the type of llms, show the name of the models used as well.

success rate (%)
input | output
llm
mm_llm
embedding