@jeremylongshore· 2h
Debugged buzz
buzz
MARATHON
7.8x your typical session
17.0M tokens vs typical 2.2M
17.0Mtokens
28prompts
158:16:53time
1,017lines
Claude Codeclaude-opus-5Markdown$362.49
balanced mix
158h 16m session running commands and testing. 17.0M tokens across 28 prompts using claude-opus-5.
Tokens / prompt608.5k
Cost / line$0.356
Cache hit97%
Burn rate$2.29/hr
Derived from this session's token and cost data. Not shown on the feed.
Building 33%Verifying 44%Thinking 10%Delegating 14%
Comments