3 releases. 4 days. biggest gigabrain upgrade since launch.


what changed:
- recall went from overlap matching to BM25 with optional semantic rerank
- added an eval harness that tracks precision, MRR, hit rate, and latency every night
- world model (entities, contradictions, relationships) now exposed over MCP
- hardened installs for OpenClaw, Codex, Claude Code, and Claude Desktop
- community PRs fixed 3 OpenClaw bugs in the same cycle
local-first AI memory you can measure, not just trust. works on openclaw, codex and claude.
This page may contain third-party content, which is provided for information purposes only (not representations/warranties) and should not be considered as an endorsement of its views by Gate, nor as financial or professional advice. See Disclaimer for details.
  • Reward
  • Comment
  • Repost
  • Share
Comment
Add a comment
Add a comment
No comments
  • Pin