ARC-AGI-2
A bird's-eye view of the benchmark: what it measures and every AI IQ chart built on it.
| # | Model | Effort | Score | Cost per task | Completion time (est.) | Weights |
|---|
About ARC-AGI-2
What ARC-AGI-2 measures
ARC-AGI-2 measures abstract reasoning: each task is a small grid puzzle that shows a few input-output examples of a hidden transformation rule, and the model must infer the rule and apply it to a new input. The puzzles are designed to be novel, so scores reflect on-the-spot reasoning rather than memorized knowledge. Every result is scored per task with a published cost per attempt, taken from the official ARC Prize leaderboard.
Reading the cost view
Each line is one model run at several reasoning-effort levels on ARC-AGI-2. Every point shows the score and cost per task at that effort, with cheaper runs on the left. Up and to the left is better; a line that stays flat while moving right is buying little extra score with the extra spend. Generation is judged among models with a published result on this benchmark, so the newest scored model in each lab's line counts as current even when a newer model has no result yet; use the Generation filter to include previous and older models.
Reading the completion-time view
Each point is a public model's ARC-AGI-2 score against its estimated completion time (time to last token) for the input and output lengths set by the sliders. Timing comes from the site's per-model speed measurements, not from the benchmark run, so every point is an estimate and labels carry a ~ prefix. The newest generation with a published result is shown by default. Up and to the left is better.
Reading the score bars
Bars rank models by the source-backed benchmark value used for this chart. Longer bars indicate higher published scores.
Reading the model table
Each row is one canonical model: the score shown is the same best published configuration the score bars use. Models with several published reasoning-effort levels expand to one sub-row per level, with the score and cost per task the source reports for that level. Completion time is the site's estimated time to last token for a 10K-token input and 1K-token output (marked ~), not a benchmark run time. The top 20 rows are shown by default; click a column heading to sort, and missing values always sort last.