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.
List-rate estimate / day
Jun 18, 2026peak $433.53Sep 11, 2026
on a plan$5,903
paid per token$1,465
per active day$111.64
13.32B
tokens
201.9M on a working day
53,387
requests
249.5K each
66
active days
of 86 tracked
32
day streak
still running
Activity
+60% vs previous 30 days
13.32B tokens over 66 active days · last 12 months
Averaging 201.9M on a day worked, with a median of 170.2M. Busiest was Aug 5, 2026 at 742.4M. 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 Sep 11, 2026. Counters only: no prompts, no project names, no file paths.
Drift
What got reached for, month by month
Each row is one month at 100% of its own tokens, so a quiet month reads the same width as a loud one and the only thing changing is the mix.
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
$5,861
cache write
$643.84
output
$591.99
fresh input
$271.25
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.
99%
of everything entering the context window came from cache instead of being paid for again
386.4×
tokens of context read per token written
249.5K
tokens in an average request
cache read
98% · 13.08B
cache write
1% · 77.0M
output
0% · 34.4M
fresh input
1% · 135.2M
Models
23 models, mostly claude-opus-5
claude-opus-5
8.28B · 62.1%
$4,915 estimated at list rates · 28,286 requests
grok-4.6
1.31B · 9.8%
$780.46 estimated at list rates · 9,472 requests
claude-sonnet-5
1.11B · 8.3%
$298.16 estimated at list rates · 3,479 requests
grok-4.5
774.5M · 5.8%
$276.97 estimated at list rates · 4,067 requests
claude-opus-4-8
482.4M · 3.6%
$389.75 estimated at list rates · 2,331 requests
gpt-5.6-luna
380.0M · 2.9%
$10.76 estimated at list rates · 1,956 requests
gpt-6-astra
331.9M · 2.5%
$502.28 estimated at list rates · 531 requests
deepseek-v4-flash
303.4M · 2.3%
$4.72 estimated at list rates · 955 requests
claude-fable-5
106.3M · 0.8%
$153.52 estimated at list rates · 488 requests
glm-5.3-flash
49.6M · 0.4%
$1.75 estimated at list rates · 438 requests
+13 more below one percent.
Tools
6 harnesses, one archive
Claude Code
10.05B · 34,928 requests · $5,777
75.4%
Grok Build
2.08B · 13,511 requests · $1,055
15.6%
OpenCode
743.4M · 3,571 requests · $21.21
5.6%
Codex
399.7M · 823 requests · $504.36
3%
Antigravity
49.3M · 547 requests · $9.79
0.4%
OpenClaw
175.4K · 7 requests
0%
Rhythm
Heaviest on Wednesday
Tokens by day of the week, in the timezone each device reported.
Mon: 1.49B over 9 days
Tue: 1.97B over 10 days
Wed: 3.36B over 9 days
Thu: 2.05B over 10 days
Fri: 2.01B over 10 days
Sat: 1.00B over 9 days
Sun: 1.45B over 9 days
MTWTFSS
By month
4 months tracked, busiest Aug 26.
Jun 26: 36.3M over 5 days
Jul 26: 2.77B over 22 days
Aug 26: 7.82B over 28 days
Sep 26: 2.70B over 11 days
JunJulAugSep
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 29 projects
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
59.9%
project 2
17.5%
project 3
8.7%
project 4
4.4%
project 5
3.1%
project 6
2.4%
project 7
1.7%
project 8
0.9%
Ordered by size, unlabelled on purpose. The top project is 59.9% 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.
