TonyBai

Sep 9, 2026 · Artificial Intelligence

YC Debunks 'Model-Only' Myth: Harness, Not Model, Sets Agent Ceiling

YC Paper Club reveals how the same Claude Opus model scores 30% on ARC-AGI bare but reaches 95% with proper Harness, and NVIDIA's AVO hits 100%, proving agent runtime—not model weights—determines the performance ceiling.