This response is so dense with numbers and special characters that it's probably about 1000 tokens. So at 1 token/s, it takes almost 17 minutes to generate this on the MacBook Pro.
But, Kimi thought for 36k tokens before writing it.
And it probably takes longer for a human to compile all that info.
But, Kimi thought for 36k tokens before writing it.