3 appearances on the backlist front page in the last 30 days.
"bothering to measure stuff that you didn't immediately need for posterity" already puts you in the upper upper percentile of anyone on earth doing serious data work
hey isn't it kind of messed up that API customers of Opus 4.7 and Opus 4.8 are paying ~1.41x as much for general english output (when measured against a consistent tokenizer baseline) vs Opus 4.6? from THE ONLY lab that's cagey about releas
anthropic pricing docs claim: "This new tokenizer may use up to 35% more tokens for the same fixed text." my measurements claim: the average inflation appears to be higher than their ceiling. and this is... general english, not arabic or so