← all stories

Arc Agi 3

Facts

Noted
a benchmark measuring AI reasoning in unknown game-like environments · 2026-08-29

Structured graph also available as JSON at /public/entities/arc-agi-3. CC BY 4.0.

All coverage

1h ago

NVIDIA Agent AVO Scores 100% on ARC-AGI-3 Where Base Model Manages 30%

NVIDIA's Agentic Variation Operators (AVO) agent system scored 100% on the public set of ARC-AGI-3, a benchmark measuring AI reasoning in unknown game-like environments. The base model 5 used inside AVO scored about 30% in a separate evaluation by ARC Prize, the benchmark's developer. AVO cleared all 183 levels across 25 environment types.