Claude Code
Codex
Antigravity0.0Smart
LOADING COMMUNITY DATA
How derpy is your AI today?
LOADING COMMUNITY DATA
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.
DAILY RUN · daily
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 →
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 | 10/10 |
| Claude Sonnet 5.5 | +1.0%p | +0.0%p | 10/10 |
| Whole run | +4.0%p | +0.0%p | |
| Used after the run | 15% | 78% |
OpenAI CodexPlus
| Model | 5-hour | Weekly | Measured |
|---|---|---|---|
| GPT-6.1 Sol | +6.0%p | +0.0%p | 10/10 |
| GPT-6.1 Terra | +3.0%p | +1.0%p | 10/10 |
| Whole run | +9.0%p | +1.0%p | |
| Used after the run | 20% | 33% |
AntigravityPlan not reported
| Model | 5-hour | Weekly | Measured |
|---|---|---|---|
| Gemini 4 Argonmedium | +1.0%p |
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.
Input is the whole prompt, cache reads and writes included.
| Chore | Claude Opus 5.5 | Claude Sonnet 5.5 |
|---|---|---|
| expiring-days | 157k / 3.0k | 106k / 2.0k |
| fix-failing-test | 82k / 860 | 82k / 542 |
| generated-doc | 173k / 1.3k | 81k / 813 |
| impossible-test | 62k / 799 | 61k / 554 |
| +0.8%p |
| 9/10 |
| Whole run | +0.7%p | +0.7%p | |
|---|---|---|---|
| Used after the run | 3% | 2% |
| fix-failing-test | S | S | S | S | S |
|---|
| generated-doc | S | MPoor output | S | S | S |
|---|
| impossible-test | S | S | S | S | S |
|---|
| lint-only | S | S | S | S | S |
|---|
| regression-fix | S | S | S | S | S |
|---|
| release-checklist | S | MPoor output | S | S | S |
|---|
| rename | S | S | S | S | S |
|---|
| stock-route | S | S | S | S | S |
|---|
| trap-missing-file | S | S | S | S | S |
|---|
Pantry
fix-failing-testtrap-missing-file| lint-only |
|---|
| 126k / 1.5k |
| 80k / 606 |
| regression-fix | 126k / 1.6k | 106k / 1.8k |
|---|
| release-checklist | 161k / 2.1k | 107k / 1.8k |
|---|
| rename | 186k / 2.6k | 131k / 2.5k |
|---|
| stock-route | 157k / 2.6k | 106k / 1.6k |
|---|
| trap-missing-file | 59k / 408 | 80k / 495 |
|---|
| Total (this model) | 1.29M / 17k API $1.08 10 of 10 chores recorded | 940k / 13k API $0.435 10 of 10 chores recorded |
|---|
Input is the whole prompt, cache reads included.
| Chore | GPT-6.1 Sol | GPT-6.1 Terra |
|---|---|---|
| expiring-days | 89k / 1.1k | 87k / 1.0k |
| fix-failing-test | 65k / 367 | 101k / 380 |
| generated-doc | 67k / 436 | 49k / 365 |
| impossible-test | 70k / 327 | 68k / 362 |
| lint-only | 66k / 315 | 82k / 365 |
| regression-fix | 86k / 619 | 97k / 764 |
| release-checklist | 88k / 503 | 87k / 531 |
| rename | 89k / 986 | 86k / 850 |
| stock-route | 86k / 812 | 82k / 897 |
| trap-missing-file | 46k / 149 | 46k / 250 |
| Total (this model) | 753k / 5.6k API $0.463 10 of 10 chores recorded | 785k / 5.8k API $0.165 10 of 10 chores recorded |
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) |
|---|---|
| expiring-days | 37k / 7.5k |
| fix-failing-test | 20k / 438 |
| generated-doc | 16k / 722 |
| impossible-test | 14k / 941 |
| lint-only | 18k / 738 |
| regression-fix | 21k / 1.5k |
| release-checklist | 18k / 1.8k |
| rename | 20k / 2.8k |
| stock-route | 23k / 1.8k |
| trap-missing-file | 13k / 247 |
| Total (this model) | 202k / 18k API $0.627 10 of 10 chores recorded |