A fun look at making coding assistants talk less, and what that actually saves in practice. The main takeaway is that shorter replies help, but most token cost still comes from the actual work: reading, reasoning, coding, and checking. Read more
A fun look at making coding assistants talk less, and what that actually saves in practice. The main takeaway is that shorter replies help, but most token cost still comes from the actual work: reading, reasoning, coding, and checking. Read more
No comments yet. Log in to reply on the Fediverse. Comments will appear here.