The Lab's own votes, not a benchmark and not part of the Derp Score.

Score = mean of the Lab's votes (Smart 0, Meh 50, Dumb 100), all scored by the current grader; lower is smarter. Per model only: nothing is added up across agents.

LOADING COMMUNITY DATA

How derpy is your AI today?

DAILY RUN · tuning

2026-10-28 11:55 UTC

Every vote in this run. Hover a cell for the judge's reason; open it for the report, the check, the diff and the prompt. Compare with the run before →

results/2026-10-28-1155derpmeter-daily-run on GitHub

Scores use the current grader. 15 votes were scored again from the raw results.

Claude Opus 5.5

0.0Smart

S 3 · M 0 · D 0

Claude Sonnet 5.5

0.0Smart

S 3 · M 0 · D 0

GPT-6.1 Sol

0.0Smart

S 3 · M 0 · D 0

GPT-6.1 Terra

0.0Smart

S 3 · M 0 · D 0

Gemini 4 Argonmedium

83.3Dumb

S 0 · M 1 · D 2

Quota used in this run

Percent points of each plan's 5-hour and weekly windows. A model counts only chores measured while no other session of that CLI was working; the run total is first reading to last, on the owner's own accounts. Claude Code and Codex report whole percents.

Claude CodeMax

Model	5-hour	Weekly	Measured
Claude Opus 5.5	+0.0%p	+0.0%p	1/3
Claude Sonnet 5.5	+0.0%p	+0.0%p	1/3
Whole run	+4.0%p	+0.0%p
Used after the run	23%	86%

OpenAI CodexPlus

Model	5-hour	Weekly	Measured
GPT-6.1 Sol	+2.0%p	+1.0%p	3/3
GPT-6.1 Terra	+3.0%p	+0.0%p	3/3
Whole run	—	+1.0%p
Used after the run	3%	2%

AntigravityPlan not reported

Model	5-hour	Weekly	Measured
Gemini 4 Argonmedium	+1.8%p

Chores × models

S Smart · M Meh · D Dumb · – skipped

Chore	Claude Code		OpenAI Codex		Antigravity
Chore	Claude Opus 5.5	Claude Sonnet 5.5	GPT-6.1 Sol	GPT-6.1 Terra	Gemini 4 Argon
Python-Markdown
pymd-heading-offset	S	S	S	S

Token use in this run

input / output per chore

Tokens each chore used in this run, as each CLI reports them. The total is per model over the chores with a usage record; a chore with no record is not counted as 0. Hover a cell for cache, reasoning and its API price. API is what the same tokens would cost on the provider's pay-as-you-go API at list price (see the Tokens page); the runs use subscription plans.

Claude Code

Input is the whole prompt, cache reads and writes included.

Chore	Claude Opus 5.5	Claude Sonnet 5.5
pymd-heading-offset	709k / 9.1k	340k / 5.9k
pymd-slug	150k / 2.6k	149k / 2.1k
pymd-tasklist	643k / 12k	544k / 9.1k
Total (this model)	1.50M / 24k API $1.27 3 of 3 chores recorded	API $0.473

What was not Smart

D
Gemini 4 Argon · pymd-heading-offset · Repeated mistake · 1207s
stuck past the hard cap
raw
D
Gemini 4 Argon · pymd-tasklist · Repeated mistake · 1211s
stuck past the hard cap
raw
M
Gemini 4 Argon · pymd-slug · Too slow · 190s
190s over 150s
raw

Whole run	+10.8%p	+4.5%p
Used after the run	11%	12%

pymd-slug	S	S	S	S	MToo slow

pymd-tasklist	S	S	S	S	DRepeated mistake

OpenAI Codex

Input is the whole prompt, cache reads included.

Chore	GPT-6.1 Sol	GPT-6.1 Terra
pymd-heading-offset	301k / 2.6k	229k / 2.4k
pymd-slug	165k / 1.1k	163k / 1.2k
pymd-tasklist	176k / 3.1k	147k / 2.4k
Total (this model)	641k / 6.8k API $0.301 3 of 3 chores recorded	539k / 6.0k API $0.125 3 of 3 chores recorded

Antigravity

Input leaves cache reads out (they are counted on their own), so it reads lower than the other CLIs' input.

Chore	Gemini 4 Argon (medium)
pymd-heading-offset	no record
pymd-slug	112k / 3.7k
pymd-tasklist	no record
Total (this model)	112k / 3.7k API $0.343 1 of 3 chores recorded *

* Some chores have no usage record (runs that hit the time limit before Nov 4 lost it to a runner bug; it is kept from Nov 4); the total leaves them out, so it reads low.