Is ARC-AGI-1 still meaningful?
Status:Saturated Checked against the report, 5 November 2026
Verdict
No. Many models now match the human score, so it cannot rank them.
Why saturated: Many models score 98.5%, and humans score 98%, so no headroom remains.[1][1] ARC-AGI-1 leaderboard data ARC Prize Primary
What “Saturated” meansKnown problems
- Models went from 50% (Apr 2025) to 90% (Dec 2025) in eight months.[2][2] Benchmark data bundle Epoch AI Primary
What to look at instead
Where it appears in the report
Chapter 1, Benchmarks
- Figure 1. Each New Benchmark Fills Up Faster Than the Last
- Figure 2. Recent benchmarks went from 50% to 90% in under a year
- Table 2. First model to cross each threshold in “Go deeper: The full saturation timeline (14 benchmarks)”
- Table 3. How each area replaced its saturated benchmarks, oldest first in “Go deeper: Lineage of replacements, by area”
- Table 5. Reasoning and knowledge benchmarks: what they measure, who maintains them, the top score on 5 November 2026 and the known flaw in “Go deeper: Reasoning and knowledge benchmarks (17)”
Sources
Copied from the chapter source lists. Each number links back to the source in its chapter.
- [1] ARC-AGI-1 leaderboard data, ARC Prize. It gives semi-private scores and the human panel result. Primary arcprize.org Chapter 1, source 35
- [2] Benchmark data bundle, Epoch AI. Dated scores (5 Nov 2026) and ceilings for most benchmarks here. Primary epoch.ai Chapter 1, source 1