AgentMemory vs MemPalace

AgentMemory

Agents

MemPalace

Agents

Tagline

AgentMemory

Open-source persistent memory runtime for AI coding agents, with hybrid retrieval and zero external dependencies.

MemPalace

Open-source, local-first AI memory system for LLM agents

Pricing

AgentMemory

Free· Free, open-source (Apache 2.0); bring your own LLM API key

MemPalace

Free· Free and open-source under the MIT license. No paid tiers; self-hosted with zero API costs once installed.

API

AgentMemory

Yes

MemPalace

Yes

Platforms

AgentMemory

cliapi

MemPalace

—

Open source

AgentMemory

Yes · Apache-2.0

MemPalace

Yes

GitHub stars

AgentMemory

29,984

checked 2026-10-29

MemPalace

—

Last GitHub push

AgentMemory

2026-10-28

MemPalace

—

First commit

AgentMemory

2026-02

MemPalace

—

Package downloads (30 d)

AgentMemory

77,138 · npm @agentmemory/agentmemory

checked 2026-11-07

MemPalace

—

Hacker News stories

AgentMemory

1

checked 2026-11-07

MemPalace

0

checked 2026-10-16

Earliest Wayback capture

AgentMemory

2026-04-19

earliest archived capture of the site, not a launch date

MemPalace

2026-04-14

earliest archived capture of the site, not a launch date

Company

AgentMemory

AgentMemory

MemPalace

—

Model used

AgentMemory

Multi-model (Claude, Gemini, MiniMax, OpenRouter)

MemPalace

Embedding backends: embedding-gemma-300m (multilingual) or all-MiniLM-L6-v2 (English). No LLM required for retrieval.

Best for

AgentMemory

Pick AgentMemory if you want a drop-in, self-hosted memory layer for your coding agent stack without running a vector DB or graph store.

MemPalace

Engineers and AI researchers building agents that need persistent, high-fidelity memory without cloud dependencies or per-token API costs, especially teams already using MCP-compatible clients like Claude Code, Cursor, or Codex CLI.

Not for

AgentMemory

Skip it if you need a hosted, multi-tenant memory service with SLAs or a production-grade datastore behind it.

MemPalace

Non-technical users who want a plug-and-play hosted chat-memory service, or teams that need a managed cloud SLA rather than a self-hosted Python package.

Editorial score

AgentMemory

7.3 / 10

MemPalace

—

Use cases

AgentMemory

agent-memorycoding-agentscontext-managementknowledge-graphmcp-server

MemPalace

Persistent memory for Claude Code and Cursor sessionsPer-agent memory isolation in multi-agent systemsMCP-exposed knowledge store for custom LLM toolsLocal RAG over long conversation historiesTemporal knowledge graph for entity-relationship trackingPrivacy-sensitive conversation logging without cloud APIsCodex CLI auto-save memory hooksResearch assistant diaries with scoped retrieval

Pros

AgentMemory

• Zero external infrastructure: single Node process, JSON on disk
• Triple-stream retrieval (BM25 + vectors + graph) with sub-20ms recall
• Native plugins for Claude Code, Cursor, Cline, Windsurf, and 15+ agents
• Apache 2.0, runs on your existing LLM subscription
• Strong benchmark numbers (95.2% R@5 on LongMemEval-S)

MemPalace

• Local-first: nothing leaves your machine unless you opt in, so no API bills or data-egress questions
• Verbatim storage avoids the fidelity loss common in summarization-based memory systems
• Structured wings/rooms/drawers layout enables scoped search rather than flat semantic lookup
• MCP server ships 36 tools and integrates with Claude Code, Codex CLI, and Cursor out of the box
• Pluggable vector backends (ChromaDB, SQLite, Milvus, Qdrant, pgvector) fit into most existing stacks
• Temporal knowledge graph with validity windows handles facts that change over time
• Strong reported benchmark: 96.6% R@5 raw on LongMemEval without any API calls

Cons

AgentMemory

• Pre-1.0 (v0.9.27), expect breaking changes
• Single-process JSON storage may not scale to team-wide deployments
• Self-hosted only; you operate the runtime
• Effectiveness depends on the LLM provider you wire in

MemPalace

• Requires a local Python 3.9+ toolchain to install cleanly (Docker mitigates this but adds its own overhead)
• Verbatim-only storage means no automatic compaction; long-running palaces will grow on disk
• Self-hosted only — no managed cloud option for teams that don't want to run infrastructure
• Small project with a narrow contributor base; support is community-driven via GitHub issues
• Non-technical users will find the MCP/CLI-first UX unfamiliar compared to hosted chat memory products

Website

AgentMemory

agent-memory.dev

MemPalace

mempalaceofficial.com

Editorial score: rule-based, 0–10, from AI-assisted profile inputs (see /methodology) — not a user rating; “—” means unscored. “Not listed” means we have no record of it, not that it is absent. GitHub figures and prices carry the date they were checked or captured; prices are shown as published, unconverted.