{"slug": "mempalace", "name": "MemPalace", "category": "ai", "category_name": "AI tools and apps", "description": "", "recommenders": 4, "commenters": 8, "warners": 2, "score": 2.207, "rising": 0.02, "first": "2026-04-07", "latest": "2026-09-07", "why": "MemPalace gives an agent excellent recall of its own conversations.", "why_is_quote": true, "has_verdict": false, "sentiment": {"recommend": 4, "neutral": 1, "compare": 1, "warn": 2}, "spark": [0, 0, 0, 0, 0, 2, 1, 0, 0, 0, 1, 0], "listed": true, "loved": [], "caveat": "", "short_quote": "MemPalace gives an agent excellent recall of its own conversations.", "kind": "", "icon": false, "domain": null, "github_repo": "github.com/milla-jovovich/mempalace", "official_url": "https://github.com/milla-jovovich/mempalace", "pricing": "", "aliases": ["MemPalace", "mempalace"], "verdict": null, "praise": [{"label": "", "summary": "", "quotes": [{"id": 793327680, "item_id": 49582980, "url": "https://news.ycombinator.com/item?id=49582980", "by": "_ink_", "date": "2026-09-06", "version": null, "stance": "recommend", "quote": "Or https://github.com/mempalace/mempalace", "excerpt": "Or https://github.com/mempalace/mempalace", "context": "", "use_case": "cross-project agent memory", "switched_from": "", "story": "OKF Agent Memory \u2013 Git-native persistent memory for AI coding agents"}, {"id": 768370339, "item_id": 48023146, "url": "https://news.ycombinator.com/item?id=48023146", "by": "hamuf", "date": "2026-05-05", "version": null, "stance": "recommend", "quote": "MemPalace gives an agent excellent recall of its own conversations.", "excerpt": "Hey HN. AIMEAT is BBS sysop culture for the AI era. Open source, MIT, npm install. Run your own node, host your friends, build apps with AI in chat, federate when you want. Back then, sysops shared files and messages between BBSes. Now you share apps, refined knowledge, and capabilities between nodes. Same culture, different payload. Quick test, paste into any AI chat that can fetch URLs: Fetch http://localhost:40050", "context": "", "use_case": "agent memory", "switched_from": "", "story": "Show HN: Aimeat, BBS-style sysop culture for the AI era"}, {"id": 762915072, "item_id": 47682192, "url": "https://news.ycombinator.com/item?id=47682192", "by": "Reebz", "date": "2026-04-07", "version": null, "stance": "recommend", "quote": "I'll be implementing mempalace for a few small agents to have memory portability that's managed locally.", "excerpt": "Many commentators are, mostly fairly, criticizing the repo due to issues raised on benchmarks. This is reasonable, however many are going further to bash the repo likely due to the authors. For me, I see a silver lining. I'll be implementing mempalace for a few small agents to have memory portability that's managed locally. I think the benchmarker who ran independent tests in GitHub issue #39 summed it up best: To be", "context": "", "use_case": "Local memory for agents", "switched_from": "", "story": "MemPalace, the highest-scoring AI memory system ever benchmarked"}, {"id": 762801872, "item_id": 47675117, "url": "https://news.ycombinator.com/item?id=47675117", "by": "denysvitali", "date": "2026-04-07", "version": null, "stance": "recommend", "quote": "this looks like a major breakthrough", "excerpt": "Yes. It's _that_ Milla Jovovich (actress known for Resident Evil). This definitely wasn't on my 2026 Bingo Card. Confirmed in this post: https://www.instagram.com/reel/DWzNnqwD2Lu This really shows how ideas are worth more than the code itself nowadays. Haven't really tried the project myself yet, but if the benchmark is correct - this looks like a major breakthrough. Even more so coming from someone which (AFAIK) is", "context": "", "use_case": "", "switched_from": "", "story": "MemPalace, the highest-scoring AI memory system ever benchmarked"}]}], "caveats": [{"label": "", "summary": "", "quotes": [{"id": 793599490, "item_id": 49599968, "url": "https://news.ycombinator.com/item?id=49599968", "by": "vshulcz", "date": "2026-09-07", "version": null, "stance": "warn", "quote": "MemPalace: ~3h mining, 2.6s query, 14 hit@1", "excerpt": "Benched cold start recently: 19k LongMemEval sessions laid down in the real ~/.claude and ~/.codex layouts, 100 questions whose answer sits in exactly one session, scored by whether that session comes back (bias: I built deja, one of the six) deja: 29s to index, 24ms query, 18/100 hit@1, 67 found@50. Plain BM25, no vectors agentmemory: 95s import, 14 hit@1, 65 found@50, plus a worker and engine on four ports MemPalac", "context": "", "use_case": "", "switched_from": "", "story": "OKF Agent Memory \u2013 Git-native persistent memory for AI coding agents"}, {"id": 763262960, "item_id": 47703935, "url": "https://news.ycombinator.com/item?id=47703935", "by": "diimdeep", "date": "2026-04-09", "version": null, "stance": "warn", "quote": "it is so dumb that it doesn't even add to the drift", "excerpt": "This is what you get combining shameless bunch of famous B-rated movie star, crypto dudes and Automatic programming hype (Claude in contibs) https://github.com/milla-jovovich/mempalace/graphs/contributors But it is so dumb that it doesn't even add to the drift towards greater Idiocracy clock values.", "context": "", "use_case": "", "switched_from": "", "story": "Creating the Futurescape for the Fifth Element [2019]"}]}], "who": [], "tags": [], "use_cases": ["cross-project agent memory", "agent memory", "ai memory", "local memory for agents"], "compared": [], "makers": [], "threads": [{"url": "https://news.ycombinator.com/item?id=47672792", "title": "MemPalace, the highest-scoring AI memory system ever benchmarked", "mentions": 3}, {"url": "https://news.ycombinator.com/item?id=49581240", "title": "OKF Agent Memory \u2013 Git-native persistent memory for AI coding agents", "mentions": 2}, {"url": "https://news.ycombinator.com/item?id=47701100", "title": "Creating the Futurescape for the Fifth Element [2019]", "mentions": 2}, {"url": "https://news.ycombinator.com/item?id=48023129", "title": "Show HN: Aimeat, BBS-style sysop culture for the AI era", "mentions": 1}], "sentiment_by_month": {"2025-11": {}, "2025-12": {}, "2026-01": {}, "2026-02": {}, "2026-03": {}, "2026-04": {"recommend": 2, "compare": 1, "neutral": 1, "warn": 1}, "2026-05": {"recommend": 1}, "2026-06": {}, "2026-07": {}, "2026-08": {}, "2026-09": {"recommend": 1, "warn": 1}, "2026-10": {}}, "versions": [], "version_unnamed": 0, "family_ref": null, "runs_on": null, "lines": [], "quotes": [{"id": 793599490, "item_id": 49599968, "url": "https://news.ycombinator.com/item?id=49599968", "by": "vshulcz", "date": "2026-09-07", "version": null, "stance": "warn", "quote": "MemPalace: ~3h mining, 2.6s query, 14 hit@1", "excerpt": "Benched cold start recently: 19k LongMemEval sessions laid down in the real ~/.claude and ~/.codex layouts, 100 questions whose answer sits in exactly one session, scored by whether that session comes back (bias: I built deja, one of the six) deja: 29s to index, 24ms query, 18/100 hit@1, 67 found@50. Plain BM25, no vectors agentmemory: 95s import, 14 hit@1, 65 found@50, plus a worker and engine on four ports MemPalac", "context": "", "use_case": "", "switched_from": "", "story": "OKF Agent Memory \u2013 Git-native persistent memory for AI coding agents"}, {"id": 793327680, "item_id": 49582980, "url": "https://news.ycombinator.com/item?id=49582980", "by": "_ink_", "date": "2026-09-06", "version": null, "stance": "recommend", "quote": "Or https://github.com/mempalace/mempalace", "excerpt": "Or https://github.com/mempalace/mempalace", "context": "", "use_case": "cross-project agent memory", "switched_from": "", "story": "OKF Agent Memory \u2013 Git-native persistent memory for AI coding agents"}, {"id": 768370339, "item_id": 48023146, "url": "https://news.ycombinator.com/item?id=48023146", "by": "hamuf", "date": "2026-05-05", "version": null, "stance": "recommend", "quote": "MemPalace gives an agent excellent recall of its own conversations.", "excerpt": "Hey HN. AIMEAT is BBS sysop culture for the AI era. Open source, MIT, npm install. Run your own node, host your friends, build apps with AI in chat, federate when you want. Back then, sysops shared files and messages between BBSes. Now you share apps, refined knowledge, and capabilities between nodes. Same culture, different payload. Quick test, paste into any AI chat that can fetch URLs: Fetch http://localhost:40050", "context": "", "use_case": "agent memory", "switched_from": "", "story": "Show HN: Aimeat, BBS-style sysop culture for the AI era"}, {"id": 763262960, "item_id": 47703935, "url": "https://news.ycombinator.com/item?id=47703935", "by": "diimdeep", "date": "2026-04-09", "version": null, "stance": "warn", "quote": "it is so dumb that it doesn't even add to the drift", "excerpt": "This is what you get combining shameless bunch of famous B-rated movie star, crypto dudes and Automatic programming hype (Claude in contibs) https://github.com/milla-jovovich/mempalace/graphs/contributors But it is so dumb that it doesn't even add to the drift towards greater Idiocracy clock values.", "context": "", "use_case": "", "switched_from": "", "story": "Creating the Futurescape for the Fifth Element [2019]"}, {"id": 763220208, "item_id": 47701263, "url": "https://news.ycombinator.com/item?id=47701263", "by": "sschueller", "date": "2026-04-09", "version": null, "stance": "neutral", "quote": "just released an AI memory called mempalace", "excerpt": "Off topic but Milla Jovovich just released an AI memory called mempalace: https://github.com/milla-jovovich/mempalace", "context": "", "use_case": "AI memory", "switched_from": "", "story": "Creating the Futurescape for the Fifth Element [2019]"}, {"id": 762915072, "item_id": 47682192, "url": "https://news.ycombinator.com/item?id=47682192", "by": "Reebz", "date": "2026-04-07", "version": null, "stance": "recommend", "quote": "I'll be implementing mempalace for a few small agents to have memory portability that's managed locally.", "excerpt": "Many commentators are, mostly fairly, criticizing the repo due to issues raised on benchmarks. This is reasonable, however many are going further to bash the repo likely due to the authors. For me, I see a silver lining. I'll be implementing mempalace for a few small agents to have memory portability that's managed locally. I think the benchmarker who ran independent tests in GitHub issue #39 summed it up best: To be", "context": "", "use_case": "Local memory for agents", "switched_from": "", "story": "MemPalace, the highest-scoring AI memory system ever benchmarked"}, {"id": 762882720, "item_id": 47680170, "url": "https://news.ycombinator.com/item?id=47680170", "by": "softcane", "date": "2026-04-07", "version": null, "stance": "compare", "quote": "Certainly not the best result compared to mempalace", "excerpt": "Certainly not the best result compared to mempalace but I was thinking something similar trough my project https://github.com/softcane/contextgc", "context": "", "use_case": "", "switched_from": "", "story": "MemPalace, the highest-scoring AI memory system ever benchmarked"}, {"id": 762801872, "item_id": 47675117, "url": "https://news.ycombinator.com/item?id=47675117", "by": "denysvitali", "date": "2026-04-07", "version": null, "stance": "recommend", "quote": "this looks like a major breakthrough", "excerpt": "Yes. It's _that_ Milla Jovovich (actress known for Resident Evil). This definitely wasn't on my 2026 Bingo Card. Confirmed in this post: https://www.instagram.com/reel/DWzNnqwD2Lu This really shows how ideas are worth more than the code itself nowadays. Haven't really tried the project myself yet, but if the benchmark is correct - this looks like a major breakthrough. Even more so coming from someone which (AFAIK) is", "context": "", "use_case": "", "switched_from": "", "story": "MemPalace, the highest-scoring AI memory system ever benchmarked"}]}