§ agents · storyline
Nvidia's AVO agent scores 100% on ARC-AGI-3 benchmark
Nvidia's AVO agent system, built on Claude Opus 5, scores 100% on all 183 levels of the ARC-AGI-3 public benchmark, up from a 30% model baseline.
Terry Chen / NVIDIA Technical Blog: Nvidia says its general-purpose coding agent system AVO scored 100% across all 25 environments in the ARC-AGI-3 public set, completing all 183 levels — The research project elevates Claude Opus 5 from a 30% model baseline to 100% as part of the complete AVO agent system, showing that system design …
§ sources1 publication · timeline below