@adam-
A lot of the competitive LLM benchmarking conversations miss the bigger picture around token efficiency.
What separates the spectators from the professionals really comes down to how you route your usage to the best model for the appropriate task.
BlockRun is definitely on to something here by taking the OpenRouter approach for Clawd.
https://x.com/bc1beat/status/2019555730475610236?s=46
0 reply
0 recast
5 reactions