@Google On each of the ARC Prize benchmarks, Carbon used more reasoning tokens than Argon.
For example, on ARC-AGI-3 Public games, Carbon at high reasoning averaged 6k reasoning tokens for its first action vs 2k for Argon.
Part of a thread · earlier post
@Google On each of the ARC Prize benchmarks, Carbon used more reasoning tokens than Argon.
For example, on ARC-AGI-3 Public games, Carbon at high reasoning averaged 6k reasoning tokens for its first action vs 2k for Argon.
ARC Prize · x.comPublished
Original title: @Google On each of the ARC Prize benchmarks, Carbon used more reasoning tokens than Argon.