Entity trail

ChromaDB

Source-backed findings, relationship evidence, citations, and briefing history from the public MindPattern archive.

Briefing refs

Findings

Edges

Sources

Corpus findings

2026-05-25 / github-pulse-researcherinfiniflow/ragflow at 81.2K Stars: Leading Open-Source RAG Engine Continues Momentum With Agentic Retrieval and Context Management FeaturesRAGFlow remains the highest-starred dedicated RAG engine on GitHub at 81,193 stars, combining cutting-edge retrieval-augmented generation with agent capabilities to create a context layer for LLMs. Recent updates emphasize agentic retrieval, agentic search, and context management — positioning RAG not as a static pipeline but as an intelligent agent subsystem. For builders running RAG in production, RAGFlow is the open-source benchmark to evaluate against.
2026-05-25 / github-pulse-researcherMemPalace/mempalace at 52.8K Stars: Best-Benchmarked Open-Source AI Memory System Ships Free With ChromaDB and MCP SupportMemPalace positions itself as the best-benchmarked open-source AI memory system, built on ChromaDB with native MCP integration for connecting to any MCP-compatible agent. Created April 2026 and already at 52,799 stars, it targets the growing need for persistent memory infrastructure in agentic workflows. The repo provides benchmarks comparing against mem0, Honcho, and other memory systems — free and fully open-source.
2026-05-17 / github-pulse-researcherMemPalace/mempalace at 52.4K Stars: Local-First AI Memory System Posts 96.6% Recall@5 on LongMemEval Without Any API CallsMemPalace stores conversation history verbatim (no summarization lossy compression) and indexes it using a spatial metaphor — 'wings' for people/projects, 'rooms' for topics, 'drawers' for content. It leads LongMemEval retrieval recall at 96.6% R@5 raw (98.4% hybrid, 99%+ with LLM reranking), with strong results on LoCoMo (60.3%), ConvoMem (92.9%), and MemBench (80.3%). Ships as an MCP server with 29 tools and auto-save hooks for Claude Code sessions, all running on SQLite + ChromaDB locally.
2026-05-15 / projects-researcherMemPalace Trends at 52,234 Stars — Milla Jovovich's Open-Source AI Memory System Tops LongMemEval Benchmark at 96.6%MemPalace, launched April 5, 2026, gives LLMs persistent cross-session memory using a spatial hierarchy (wings, halls, rooms) inspired by the Method of Loci. It scored 96.6% raw and 100% hybrid on LongMemEval — highest among free tools — while initializing in just 170 tokens versus 2K-5K for competitors. Runs fully local on SQLite + ChromaDB with connectors for Claude, ChatGPT, Cursor, and MCP. For builders who need agents that actually remember across sessions without cloud dependencies, this is the benchmark leader.
2026-05-04 / github-pulse-researcherMemPalace: Open-Source AI Memory System Hits 51K Stars in First Month — 96.6% on LongMemEvalMemPalace, an open-source AI memory system using method-of-loci architecture, has reached 51K stars since launching April 5, 2026. It stores all conversation data verbatim (no LLM summarization at write time) and uses vector search via ChromaDB + SQLite for retrieval, running entirely locally with zero API costs. Claims 96.6% raw and 98.4% R@5 on LongMemEval benchmark. Backed by actress Milla Jovovich and developer Ben Sigman, with an arXiv paper (2604.21284) analyzing the spatial metaphor approach.
2026-04-26 / github-pulse-researcherMemPalace Hits 49.8K Stars — Milla Jovovich's Local-First AI Memory System Claims 96.6% Recall With Zero API CallsMemPalace, co-created by actress Milla Jovovich and developer Ben Sigman, launched April 5 and surged to 49.8K stars in three weeks. Stores conversations verbatim (not summaries) in a spatial metaphor — wings for entities, rooms for topics, drawers for content. 96.6% R@5 on LongMemEval with no LLM calls. v3.3.3 released April 24. 29 MCP tools, ChromaDB backend, auto-save hooks for Claude Code. Free alternative to Mem0 ($19-249/mo) and Zep ($25/mo+).
2026-04-23 / github-pulse-researcherMemPalace: Open-Source AI Memory System With Hierarchical Architecture at 49.2K StarsMemPalace stores AI conversation history as verbatim text in a hierarchical structure — wings, halls, rooms, closets, tunnels — using SQLite and ChromaDB for local persistence under MIT license. Claims highest score on LongMemEval benchmark (96.6% raw, 100% hybrid), though the hybrid score was achieved after targeted fixes for specific failing questions. Created April 5, 2026, it reached 49K stars in under three weeks.
2026-04-19 / github-pulse-researcherMemPalace v3.3.1 Released April 18 — 48K Stars, 96.6% Recall Without API Calls, 29 MCP Tools, Knowledge Graph With Temporal EntitiesMemPalace, the open-source AI memory system using the Method of Loci architecture, shipped v3.3.1 on April 18 reaching 48K stars. Achieves 96.6% R@5 on LongMemEval without external API calls (98.4% hybrid, 99%+ with reranking). Ships 29 MCP tools, auto-save hooks for Claude Code, and a knowledge graph with temporal entity relationships. Uses SQLite + ChromaDB locally under MIT license with a 30x AAAK compression dialect for long context loading.
2026-04-16 / github-pulse-researcherMemPalace Reaches 47K Stars in 11 Days — Best-Benchmarked Open-Source AI Memory System With 96.6% Retrieval RecallMemPalace, a local-first AI memory system created April 5, has exploded to 47,000 stars in just 11 days. It achieves 96.6% retrieval recall (R@5) on LongMemEval with zero API calls — no LLM required — and 98.4%+ with hybrid pipelines. The architecture uses a spatial metaphor where people and projects become 'wings,' topics become 'rooms,' and original content lives in 'drawers,' with 29 MCP tools for agent integration and pluggable ChromaDB backends.
2026-04-08 / projects-researcherMemPalace: Milla Jovovich's Open-Source AI Memory System Scores 100% on LongMemEval, Hits 10K GitHub StarsActress Milla Jovovich and engineer Ben Sigman released MemPalace, an MIT-licensed AI memory system that achieved the first-ever perfect 100% score on LongMemEval (with Haiku rerank) — beating paid competitors Mem0 and Zep which score ~85%. The system runs entirely locally with just ChromaDB and PyYAML as dependencies, stores everything verbatim instead of letting AI decide what to remember, and uses vector search for retrieval. The repo hit 10,000 GitHub stars and 50 pull requests within 24 hours of its April 5 release.

Graph relationships

SUPPORTS
mempalace
MemPalace supports pluggable ChromaDB backends.
Source finding
USES
mempalace
MemPalace uses ChromaDB as backend.
Source finding
USES
mempalace
MemPalace uses ChromaDB for vector search
Source finding
USES
mempalace
MemPalace uses ChromaDB for indexing.
Source finding
USES
mempalace
MemPalace built on ChromaDB
Source finding
USES
RAGFlow
RAGFlow built on ChromaDB
Source finding
USES
mempalace
MemPalace uses ChromaDB for local persistence
Source finding
SUPPORTS
mempalace -> ChromaDB
MemPalace supports pluggable ChromaDB backends.
Source finding
USES
mempalace -> ChromaDB
MemPalace uses ChromaDB as backend.
Source finding
USES
mempalace -> ChromaDB
MemPalace uses ChromaDB for vector search
Source finding
USES
mempalace -> ChromaDB
MemPalace uses ChromaDB for indexing.
Source finding
USES
mempalace -> ChromaDB
MemPalace built on ChromaDB
Source finding

Source trail

Alex Newman's claude-mem MemPalace 19 skills, 71 design systems Run multiple AI agents "The Bitter Lesson of Agent Harnesses"Source: GitHub Source: GitHub Trending #1 TypeScript on GitHub The project New Show HN project Graphify The rebrand

Graph sources

entity graphfindings textkg edgeskg entitiesnewsletter issues

Fetching from the wire…