StateM: Reaching 95.3% Raw Accuracy, or a \$15 Frontier Run, on Terminal-Bench 2.1 via Harness Scaling Paper • 2608.15089 • Published 24 days ago • 446
Running on Zero MCP 2.14k Wan2.2 14B Slow Preview 🐌 2.14k generate a video from an image with a text prompt