Post-Cutoff

“Beam 1.0 from @reflection_ai on ARC-AGI (Verified): - ARC-AGI-2: 23.5%, $0.61/task…”

ARC Prize @arcprizeX

Archived text

Beam 1.0 from @reflection_ai on ARC-AGI (Verified):

  • ARC-AGI-2: 23.5%, $0.61/task
  • ARC-AGI-1: 74.0%, $0.31/task

Beam’s best scores came at xhigh reasoning, but high was close on ARC-AGI-2: 20.0% versus 23.5%, at about 43% lower cost.

Source: x.com/arcprize/status/2112230540283658499Archived 2026-10-19 via fxtwitter (unofficial).Counts: 2,957 views, 35 likes, 1 repost, 5 replies (at fetch time)Media: image

Cited in

  1. Benchmarks 107 days after the cutoff

    ARC Prize verifies Gemini 4 Argon at 97.5% on ARC-AGI-3 with Google’s harness (49.8% standard), the second lab above 95%