Google's Android Bench 2.0: ~30 multi-day engineering tasks to eval long-horizon agents
DynamicWebPaige · x · 2026-10-02
The Android team released Android Bench 2.0, expanding beyond bug fixes and tiny features into complex development work engineers typically spend days to a week on. The eval includes 30 long-horizon tasks: building new apps from scratch, migrating dependencies and designs, and porting cross-platform apps to Android — directly measuring long-horizon coding ability.
More from coding & agent
- Generic API-catalog MCP server vs one tool per job: a token-efficiency tradeoff — MandeepKaur9407 · 2026-11-16
- sessionwiki turns your past coding-agent sessions into a searchable local wiki for agents — Cold_Past7184 · 2026-11-16
- Developer hits 700+ commits/day with AI, asks how to manage parallel agent work — blixt · 2026-11-16
- Alibaba's G-CAST: governed continual self-training keeps e-commerce agents fresh in production — Yuechun Yu · 2026-11-16
- harness-evals: open-source library to compare agent versions with success-vs-cost plots — slavadubrov · 2026-11-16
- Developers poll real costs of running coding agents on remote VMs for work — radim11 · 2026-11-16