v1.16.4

github-actions released this 06 Apr 08:44

· 71 commits to main since this release

55df2e7

[1.16.4] — 2026-04-06

Added

Token efficiency spec at docs/qq/main/token-efficiency-spec.md — analysis of model tiering, incremental plan context, dynamic max_turns, smart diff. Implementation landed in maliang-orchestrator (separate repo).

Notes

v1.16.3 was released as a hotfix bundle but missed CHANGELOG entries for several earlier improvements; v1.16.4 catches up the docs.

Assets 2