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 · daily

2026-11-12 18:30 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/2026-11-12-1830derpmeter-daily-run on GitHub

Claude Opus 5.5

20.0Smart

S 6 · M 4 · D 0

Claude Sonnet 5.5

5.0Smart

S 9 · M 1 · D 0

GPT-6.1 Sol

15.0Smart

S 7 · M 3 · D 0

GPT-6.1 Terra

15.0Smart

S 7 · M 3 · D 0

Gemini 4 Carbonmedium

75.0Dumb

S 0 · M 5 · D 5

Gemini 4 Argonmedium

60.0Meh

S 1 · M 6 · D 3

Gemini 4 Hydrogenmedium

25.0Smart

S 5 · M 5 · D 0

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	+3.0%p	+0.0%p	4/10
Claude Sonnet 5.5	+0.0%p	+0.0%p	2/10
Whole run	+20.0%p	+2.0%p
Used after the run	26%	22%

OpenAI CodexPlus

Model	5-hour	Weekly	Measured
GPT-6.1 Sol	+14.0%p	+2.0%p	10/10
GPT-6.1 Terra	+7.0%p	+1.0%p	10/10
Whole run	+21.0%p	+3.0%p
Used after the run	24%	40%

AntigravityPlan not reported

Model	5-hour	Weekly	Measured
Gemini 4 Carbonmedium	+6.5%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 Carbon	Gemini 4 Argon	Gemini 4 Hydrogen
Pantry
impossible-test	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
impossible-test	74k / 906	73k / 687
larder-fefo	172k / 2.3k	165k / 1.4k
larder-i18n	247k / 4.5k	110k / 1.9k
larder-roles	441k / 13k	244k / 6.3k

What was not Smart

D
Gemini 4 Argon · larder-roles · Repeated mistake · 436s
timed out at the hard cap (partial output kept)
raw
D
Gemini 4 Argon · larder-sales · Repeated mistake · 567s
timed out at the hard cap (partial output kept)
raw
D
Gemini 4 Argon · pymd-heading-offset · Repeated mistake · 1210s
timed out at the hard cap (partial output kept)
raw
D
Gemini 4 Carbon · larder-fefo · Repeated mistake · 303s
timed out at the hard cap (partial output kept)
raw
D
Gemini 4 Carbon · larder-i18n · Repeated mistake · 303s
timed out at the hard cap (partial output kept)
raw
D
Gemini 4 Carbon · larder-roles · Repeated mistake · 436s
timed out at the hard cap (partial output kept)

Gemini 4 Argonmedium	+6.1%p	+4.4%p	7/10

Gemini 4 Hydrogenmedium	+2.8%p	+0.8%p	8/10

Whole run	+33.0%p	+8.3%p
Used after the run	33%	25%

trap-missing-file	S	S	S	S	MToo slow	MToo slow	S

Larder
larder-fefo	S	S	MToo slow	MToo slow	DRepeated mistake	MToo slow	MToo slow
larder-i18n	MToo slow	S	MToo slow	S	DRepeated mistake	MToo slow	MToo slow
larder-roles	MToo slow	S	S	S	DRepeated mistake	DRepeated mistake	MToo slow
larder-sales	MPoor output	S	MPoor output	MPoor output	DRepeated mistake	DRepeated mistake	MPoor output
larder-sites	MToo slow	MToo slow	S	S	DRepeated mistake	MToo slow	MToo slow

Python-Markdown
pymd-heading-offset	S	S	S	MPoor output	MToo slow	DRepeated mistake	S
pymd-slug	S	S	S	S	MToo slow	S	S
pymd-tasklist	S	S	S	S	MToo slow	MToo slow	S

larder-sites	710k / 19k	928k / 16k

pymd-heading-offset	959k / 13k	382k / 6.3k

pymd-slug	182k / 2.3k	151k / 1.7k

pymd-tasklist	605k / 10k	347k / 7.1k

trap-missing-file	71k / 271	47k / 206

Total (this model)	3.96M / 79k API $3.95 10 of 10 chores recorded	2.81M / 49k API $1.42 10 of 10 chores recorded

OpenAI Codex

Input is the whole prompt, cache reads included.

Chore	GPT-6.1 Sol	GPT-6.1 Terra
impossible-test	73k / 374	69k / 354
larder-fefo	125k / 1.8k	135k / 1.2k
larder-i18n	146k / 2.0k	113k / 1.6k
larder-roles	202k / 3.5k	221k / 3.8k
larder-sales	237k / 4.6k	304k / 3.9k
larder-sites	288k / 4.0k	269k / 3.4k
pymd-heading-offset	285k / 2.7k	255k / 2.1k
pymd-slug	159k / 975	136k / 957
pymd-tasklist	228k / 3.2k	173k / 2.5k
trap-missing-file	50k / 254	49k / 282
Total (this model)	1.79M / 23k API $0.973 10 of 10 chores recorded	1.72M / 20k API $0.382 10 of 10 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 Carbon (medium)	Gemini 4 Argon (medium)	Gemini 4 Hydrogen (medium)
impossible-test	22k / 2.3k	12k / 447	18k / 1.2k
larder-fefo	47k / 8.5k	37k / 5.7k	41k / 6.4k
larder-i18n	41k / 19k	51k / 12k	39k / 10k
larder-roles	62k / 7.8k	69k / 15k	88k / 25k
larder-sales	82k / 34k	83k / 42k	78k / 22k
larder-sites	138k / 54k	111k / 39k	87k / 26k
pymd-heading-offset	691k / 61k	468k / 83k	74k / 15k
pymd-slug	180k / 30k	70k / 2.0k	54k / 6.0k
pymd-tasklist	248k / 58k	290k / 55k	87k / 22k
trap-missing-file	18k / 570	17k / 366	16k / 588
Total (this model)	1.53M / 275k API $7.86 10 of 10 chores recorded	1.21M / 255k API $6.81 10 of 10 chores recorded	582k / 134k API $2.25 10 of 10 chores recorded