promptdojo_

Read the editor. The same agent, 100 calls. Stable content per call: 8000 tokens. Variable content: 200 tokens. Compute the uncached cost first. The cached cost should be much lower because reads are 0.1× input rate.

What two lines does the script print (rounded to 2 decimals)?

read, then continue.