NVIDIA's AVO framework boosts Claude Opus to 100% on ARC-AGI benchmark
imjustnewatai · x · 2026-08-22
NVIDIA demonstrated a significant AI jump driven by the software wrapping the model. While Claude Opus scored around 30% on the ARC Prize, NVIDIA achieved 100% RHAE (183/183 levels) using the AVO framework. AVO acts as a harness with persistent memory, tools, execution feedback, and a supervisor, allowing discoveries to compound across context windows. Interestingly, the same AVO loop previously optimized B200 attention kernels, beating FlashAttention-4 by up to 10.5%.
Related event: NVIDIA's AVO Agent Hits 100% on ARC-AGI-3 Public Set(22 posts)→
More from coding & agent
- "Bribe coding": getting Claude to work by bribing it, and running three agents in parallel — jia_seed · 2026-10-16
- Vercel Connect adds MCP Events in public beta for event-driven agents — cramforce · 2026-10-16
- PE operating partner shares real agent deployments: 59/60 accuracy at $250/month — alex_verem · 2026-10-16
- Dev: live voice + tool-calling + image gen now combine into a workable 'Her'-like experience — andreisavu · 2026-10-16
- Kinex: a harness that connects coding agents to any physical robot for scalable learning — daibond_alpha · 2026-10-16
- Developer rebuilds GitHub in 4 days: Keel makes agents declare intent before changing code — threepointone · 2026-10-16