Author(s): yooiken
Originally published on Towards AI.
I have become less interested in whether a cheap model can “browse the web” and more interested in whether it can run a boring workflow correctly every morning.
That is a different problem.
Most low-cost or free LLMs fail at web automation because the model has to do too much at once. It has to understand the goal, inspect the page, decide where to click, recover from layout changes, parse the result, and then write something useful. One weak link ruins the whole run.
