Fetching from the wire…
OSS2026-07-30 · source-backed
MemPalace/mempalace (57,888 stars) stores conversation history as verbatim text and retrieves with semantic search, explicitly refusing to summarize, extract, or paraphrase. The index is structured into wings (people/projects), rooms (topics), and drawers (original content) so searches can be scoped rather than run flat. Pluggable backend, currently defaulting to ChromaDB. The repo now leads with a caution that impostor sites on .tech, .net, and other .com variants may distribute malware. That's a new failure mode for high-star AI infrastructure and worth checking before you curl | sh anything.
Each link below shares sources, entities, or timing with this story.
MemPalace (57,821 stars, v3.6.0) reports 96.6% raw recall@5 on LongMemEval with no LLM required, 98.4% with hybrid v4 on a held-out 450 questions, LoCoMo R@10 rising 60.3% → 88.9%, ConvoMem 92.9%, MemBench 80.3%, while explicitly refusing head-to-head comparison against Mem0,...
Created by Milla Jovovich and developer Ben Sigman using Claude Code, the MIT-licensed system uses "palace architecture" with 30x lossless compression and runs entirely offline via MCP. 50.5K stars in under a month. The celebrity angle is noise. The benchmark claim (96.6% raw,...
The submission titled "CEO fired developers to make room for AI. Developers create open source AI CEO" pointed at SenteLabsAI/OpenExecutive, an Apache-2.0 FastAPI and Next.js 15 app running eight specialist Claude agents (CSO, CFO, CHRO, General Counsel, COO, CMO, CPO, Board C...
MemPalace (~53.6K stars) claims best-benchmarked, against mem0 (~57.8K) and claude-mem (~80.8K) (GitHub). Agent memory went from experimental nicety to a competitive subcategory with published benchmarks. Pair this with the local-first angle: Mnemo offers a Rust + SQLite + pet...
MemPalace at 58,086 stars and 7,468 forks calls itself "the best-benchmarked open-source AI memory system," TencentDB-Agent-Memory is at 14,642 and climbing 1,891/day, and zszz3/AgentRecall added 278 this week to reach 699. Each attacks a different axis: benchmark leadership,...
MemPalace bills itself as "the best-benchmarked open-source AI memory system," and at 57,551 stars it's a leading local-first memory layer. Its headline 100% LoCoMo score across all five question types was achieved at top_k=50, which effectively retrieves the entire corpus, me...
MindPattern daily
One email a day at 7 AM. Sources and a take on every story. Unsubscribe anytime.