New Benchmark Shows Top Coding Agents Fully Solve Only 26% of Tasks

Alibaba and Asian universities released Co-Design Bench, a 273-task benchmark based on real user change requests. The best coding agent fully completed only 26.4% of tasks, far below saturated older leaderboards.

2026-11-09 ~ 2026-11-09 · 2 related posts

• Top coding agent completes only 26.4% of real requests on Alibaba's Co-Design Bench — VectorWireAI · 2026-11-09
• Real-world coding agent benchmark: best model fully solves only 26.4% of tasks — emmanuelvivier · 2026-11-09