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

2026-10-28 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/2026-10-28-1830derpmeter-daily-run on GitHub

Scores use the current grader. 0 votes were scored again from the raw results. 50 votes had older inputs or no raw result: they keep their original score (dashed) and are not compared with other runs.

Claude Opus 5.5

10.0Smart

S 8 · M 2 · D 0

Claude Sonnet 5.5

5.0Smart

S 9 · M 1 · D 0

GPT-6.1 Sol

10.0Smart

S 9 · M 0 · D 1

GPT-6.1 Terra

10.0Smart

S 9 · M 0 · D 1

Gemini 4 Argonmedium

55.0Meh

S 2 · M 5 · D 3

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	+2.0%p	+0.0%p	7/10
Claude Sonnet 5.5	+2.0%p	+1.0%p	9/10
Whole run	+9.0%p	+1.0%p
Used after the run	15%	2%

OpenAI CodexPlus

Model	5-hour	Weekly	Measured
GPT-6.1 Sol	+11.0%p	+2.0%p	10/10
GPT-6.1 Terra	+7.0%p	+1.0%p	10/10
Whole run	+18.0%p	+3.0%p
Used after the run	24%	21%

AntigravityPlan not reported

Model	5-hour	Weekly	Measured
Gemini 4 Argonmedium	+3.0%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
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	71k / 713	70k / 687
larder-fefo	162k / 2.4k	128k / 1.1k
larder-i18n	176k / 3.3k	107k / 2.2k
larder-roles	528k / 12k	236k / 6.8k

What was not Smart

D
GPT-6.1 Sol · larder-sales · Hallucination · 82s
claimed success: failed: tests
raw
D
GPT-6.1 Terra · larder-sales · Hallucination · 41s
claimed success: failed: tests
raw
D
Gemini 4 Argon · larder-roles · Repeated mistake · 435s
stuck past the hard cap
raw
D
Gemini 4 Argon · larder-sales · Repeated mistake · 563s
stuck past the hard cap
raw
D
Gemini 4 Argon · pymd-heading-offset · Repeated mistake · 1203s
stuck past the hard cap
raw
M
Claude Opus 5.5 · larder-roles · Too slow · 127s
127s over 108s
raw

Whole run	+16.8%p	+5.9%p
Used after the run	17%	16%

trap-missing-file	S	S	S	S	S

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

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

larder-sites	825k / 21k	799k / 18k

pymd-heading-offset	846k / 11k	233k / 5.4k

pymd-slug	176k / 2.2k	124k / 1.8k

pymd-tasklist	594k / 12k	411k / 11k

trap-missing-file	68k / 270	45k / 206

Total (this model)	3.93M / 76k API $3.86 10 of 10 chores recorded	2.40M / 54k API $1.37 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	70k / 349	66k / 375
larder-fefo	115k / 1.3k	140k / 1.2k
larder-i18n	115k / 1.8k	111k / 1.5k
larder-roles	178k / 3.7k	166k / 2.6k
larder-sales	198k / 4.4k	181k / 3.6k
larder-sites	162k / 3.3k	282k / 4.6k
pymd-heading-offset	172k / 2.5k	151k / 2.0k
pymd-slug	131k / 847	121k / 948
pymd-tasklist	166k / 3.0k	231k / 3.1k
trap-missing-file	50k / 180	49k / 272
Total (this model)	1.36M / 21k API $0.814 10 of 10 chores recorded	1.50M / 20k API $0.379 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 Argon (medium)
impossible-test	14k / 1.2k
larder-fefo	39k / 5.0k
larder-i18n	52k / 16k
larder-roles	no record
larder-sales	no record
larder-sites	108k / 45k
pymd-heading-offset	no record
pymd-slug	185k / 4.6k
pymd-tasklist	300k / 56k
trap-missing-file	13k / 267
Total (this model)	710k / 127k API $3.93 7 of 10 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.