Average API cost per task on ARC-AGI v1 — measures how much compute a model needs to attempt each reasoning puzzle.
Pairs performance with per-task spend, enabling cost-adjusted comparisons for deployment decisions; highlights when small accuracy gains require disproportionate budget increases.
Costs depend on provider pricing and inference policy at evaluation time, so historical values can drift; retry budgets and sampling settings can materially change measured cost per task.
Lower is better. Cost in USD per task.
Costs depend on provider pricing and inference policy at evaluation time, so historical values can drift.
| # | Model | Score |
|---|---|---|
| 1 | 0.0 | |
| 2 | 0.0 | |
| 3 | 0.0 | |
| 4 | 0.0 | |
| 5 | 0.0 | |
| 6 | 0.0 | |
| 7 | 0.0 | |
| 8 | 0.0 | |
| 9 | 0.0 | |
| 10 | 0.0 | |
| 11 | 0.0 | |
| 12 | 0.0 | |
| 13 | 0.0 | |
| 14 | 0.0 | |
| 15 | 0.0 | |
| 16 | 0.0 | |
| 17 | 0.0 | |
| 18 | 0.0 | |
| 19 | 0.0 | |
| 20 | 0.0 | |
| 21 | 0.0 | |
| 22 | 0.0 | |
| 23 | 0.0 | |
| 24 | 0.0 | |
| 25 | 0.0 | |
| 26 | 0.0 | |
| 27 | 0.0 | |
| 28 | 0.0 | |
| 29 | 0.0 | |
| 30 | 0.0 | |
| 31 | 0.0 | |
| 32 | 0.0 | |
| 33 | 0.0 | |
| 34 | 0.0 | |
| 35 | 0.0 | |
| 36 | 0.0 | |
| 37 | 0.0 | |
| 38 | 0.0 | |
| 39 | 0.0 | |
| 40 | 0.0 | |
| 41 | 0.0 | |
| 42 | Gemini 2.5 Flash-Lite Preview (Sep '25) (Non-reasoning) source benchpress Evidence observed Apr 13, 2026 | 0.0 |
| 43 | 0.0 | |
| 44 | 0.0 | |
| 45 | 0.0 | |
| 46 | 0.0 | |
| 47 | 0.0 | |
| 48 | 0.0 | |
| 49 | 0.0 | |
| 50 | 0.0 |