Intel / article

NVIDIA AVO Reaches 100% on ARC-AGI-3, Demonstrating a Frontier-Level General-Purpose Architecture for Long-Horizon Autonomous Agents

Source: developer.nvidia.com
Author: Tanya Lenz
Date: Published Aug 21, 2026

Context

NVIDIA reports that its AVO completed all 183 levels in ARC-AGI-3's 25 public environments using Anthropic's Claude Opus 5. These are unfamiliar interactive puzzles whose rules and goals the agent must infer. AVO preserves results and working knowledge across extended runs, while a supervisor can redirect stalled exploration. Research account.

The result concerns a complete agent system on the public set. Differences in reasoning settings and implementation prevent attributing its gain over a model baseline to memory alone. It complements earlier coverage of harness settings: OpenAI separately reported improvements from retaining reasoning and summarizing long histories. Both accounts make the surrounding software relevant to interpreting a model's score.

Terms in this piece · Glossary

AI agent — An AI system that doesn't just answer once but works toward a goal in a loop — taking actions, reading the results, and deciding what to do next.
agent harness — The scaffolding around a model that turns it into a working agent — the loop, the tools it can call, and the rules for when to stop.
tool use — A model's ability to call external functions — run code, search the web, edit files — instead of only generating text.

Read the source developer.nvidia.com