c/programmer_humor@programming.dev · sanitation@lemmy.today · 2 Sep 2026 23:30 New compression technique check post at origin
Saves input tokens and uses more output because of OCR, and output is priced way higher.
It’s an accounting trick. It uses a different budget…
Also if you’re doing anything multi-prompt, that output is re-used as (luckily cached) input tokens for the next prompt anyway lol