Arbor, an open-source AI framework from Renmin University and Microsoft Research, outperforms Codex and Claude Code by 2.5x across optimization benchmarks.

Arbor separates strategy from execution using isolated git worktrees, so engineering teams can finally trace which optimization actually moved the needle.

Arbor, an open-source AI framework from Renmin University and Microsoft Research, outperforms Codex and Claude Code by 2.5x across optimization benchmarks.

A new framework, Arbor, they claim, preserves hypotheses, experiments, and lessons learned across long-running research tasks, delivering 2.5x better performance than other models…