What these tokens would have cost billed per token at each model's published price. It is the one figure that compares across people on different plans.
on a plan$0.90
paid per token$0.00
per active day$0.90
4.4M
tokens
4.4M on a working day
9
requests
488.8K each
1
active days
not a day missed
1
day streak
still running
Activity
4.4M tokens over 1 active days · last 12 months
Averaging 4.4M on a day worked, with a median of 4.4M. Busiest was Aug 28, 2026 at 4.4M. Cells are shaded by what the day was worth at list rates, not by token count. Hover a day for the models and harnesses behind it.
less spendmore spend
Last 12 months in full on this profile, last synced Aug 28, 2026. Counters only: no prompts, no project names, no file paths.
Where it goes
$0.90, split four ways
Output is a fraction of the tokens and usually most of the money, because it is priced at several times the rate of anything read back out of cache.
cache read
$0.88
output
$0.01
cache write
$0.01
fresh input
$0.00
Every model in this archive has a known rate.
Cache
Reuse, not volume
A raw token total mostly measures how much context got re-read. The ratios underneath it are the part that says anything about how the work was done.
100%
of everything entering the context window came from cache instead of being paid for again
3414.2×
tokens of context read per token written
488.8K
tokens in an average request
cache read
100% · 4.4M
cache write
0% · 3.0K
output
0% · 1.3K
fresh input
0% · 18
Models
1 models, mostly claude-sonnet-5
claude-sonnet-5
4.4M · 100%
$0.90 estimated at list rates · 9 requests
Tools
1 harnesses, one archive
Claude Code
4.4M · 9 requests · $0.90
100%
Rhythm
Heaviest on Friday
Tokens by day of the week, in the timezone each device reported.
Mon: 0 over 0 days
Tue: 0 over 0 days
Wed: 0 over 0 days
Thu: 0 over 0 days
Fri: 4.4M over 1 day
Sat: 0 over 0 days
Sun: 0 over 0 days
MTWTFSS
Hour of day is not in the sync payload yet, so the rhythm here stops at the day. See docs/tokenstat-sync-cli.md for what the CLI would need to send.
Spread
Across 1 project
Project names stay on the device. What reaches us is an opaque hash we can count but never read, so a profile can show how the work was spread and never which work it was.
project 1
100%
Ordered by size, unlabelled on purpose. The top project is 100% of everything.
How the counts were read
Some harnesses report token counts exactly. Others have to be derived from what they write to disk. The archive keeps that distinction rather than flattening it, so a number that was inferred never passes itself off as a measurement.
