Measurement tasks also consume tokens, so I can't easily attempt them. In my subjective view, it feels like it's improved by about 20-30% compared to Opus 4.8. (The impressive aspects of performance are less than expected.......somewhat....)
It's included in the plan through June 22nd, and I don't know what will happen after that. If the cost is over $200, it seems like it would be quite a burden. (I'll probably go with it though...-_;;;)
Thanks to tokens running out quickly from this morning, I was able to go exercise early today.
