Igor Lins e Silva
89904ed03f
Five findings from the automated review, fixed with targeted tests where behavior changed: 1. Transformation Protocol (transforms.py). The registry mixed a bytes-to-str transform (utf8_replace_invalid) with str-to-str transforms under a single Callable[..., str] type, misleading static type checkers and adapter authors. Introduced a Transformation Protocol with __call__(data: bytes|str) -> str and retyped the registry + get_transformation return. 2. Drawer-id collision risk (context.py). Switched _build_drawer_id from sha1[:16]=64 bits to sha256[:24]=96 bits. 64 bits sits uncomfortably close to the birthday bound for palace-sized corpora; 96 bits keeps the collision probability negligible while preserving the existing <prefix>_<chunk> layout adapters rely on. 3. Fresh-schema KG columns (knowledge_graph.py). source_drawer_id and adapter_name now live in the canonical CREATE TABLE so new palaces don't take an ALTER round-trip on first open. _migrate_schema stays for legacy palaces (SQLite has no ADD COLUMN IF NOT EXISTS, so PRAGMA introspection is still needed there). 4. Identity-shim comment (transforms.py). Comment said the adapter-specific transforms "raise if invoked without adapter context" but they return the input unchanged. Updated the comment to match the actual identity- shim behavior Copilot suggested. 5. Test docstring (test_sources.py). Comment mentioned default_factory=list but SourceRef.options uses default_factory=dict. Corrected. Tests: 1020 passed (up from 1018), +2 new tests for the sha256 id shape and the fresh-schema column presence on new palaces.
Caution
Scam alert. The only official sources for MemPalace are this GitHub repository, the PyPI package, and the docs site at mempalaceofficial.com. Any other domain — including
mempalace.tech— is an impostor and may distribute malware. Details and timeline: docs/HISTORY.md.
MemPalace
Local-first AI memory. Verbatim storage, pluggable backend, 96.6% R@5 raw on LongMemEval — zero API calls.
What it is
MemPalace stores your conversation history as verbatim text and retrieves it with semantic search. It does not summarize, extract, or paraphrase. The index is structured — people and projects become wings, topics become rooms, and original content lives in drawers — so searches can be scoped rather than run against a flat corpus.
The retrieval layer is pluggable. The current default is ChromaDB; the
interface is defined in mempalace/backends/base.py
and alternative backends can be dropped in without touching the rest of
the system.
Nothing leaves your machine unless you opt in.
Architecture, concepts, and mining flows: mempalaceofficial.com/concepts/the-palace.
Install
pip install mempalace
mempalace init ~/projects/myapp
Quickstart
# Mine content into the palace
mempalace mine ~/projects/myapp # project files
mempalace mine ~/chats/ --mode convos # conversation exports
# Search
mempalace search "why did we switch to GraphQL"
# Load context for a new session
mempalace wake-up
For Claude Code, Gemini CLI, MCP-compatible tools, and local models, see mempalaceofficial.com/guide/getting-started.
Benchmarks
All numbers below are reproducible from this repository with the commands
in benchmarks/BENCHMARKS.md. Full
per-question result files are committed under benchmarks/results_*.
LongMemEval — retrieval recall (R@5, 500 questions):
| Mode | R@5 | LLM required |
|---|---|---|
| Raw (semantic search, no heuristics, no LLM) | 96.6% | None |
| Hybrid v4, held-out 450q (tuned on 50 dev, not seen during training) | 98.4% | None |
| Hybrid v4 + LLM rerank (full 500) | ≥99% | Any capable model |
The raw 96.6% requires no API key, no cloud, and no LLM at any stage. The hybrid pipeline adds keyword boosting, temporal-proximity boosting, and preference-pattern extraction; the held-out 98.4% is the honest generalisable figure.
The rerank pipeline promotes the best candidate out of the top-20
retrieved sessions using an LLM reader. It works with any reasonably
capable model — we have reproduced it with Claude Haiku, Claude Sonnet,
and minimax-m2.7 via Ollama Cloud (no Anthropic dependency). The gap
between raw and reranked is model-agnostic; we do not headline a "100%"
number because the last 0.6% was reached by inspecting specific wrong
answers, which benchmarks/BENCHMARKS.md flags as teaching to the test.
Other benchmarks (full results in benchmarks/BENCHMARKS.md):
| Benchmark | Metric | Score | Notes |
|---|---|---|---|
| LoCoMo (session, top-10, no rerank) | R@10 | 60.3% | 1,986 questions |
| LoCoMo (hybrid v5, top-10, no rerank) | R@10 | 88.9% | Same set |
| ConvoMem (all categories, 250 items) | Avg recall | 92.9% | 50 per category |
| MemBench (ACL 2025, 8,500 items) | R@5 | 80.3% | All categories |
We deliberately do not include a side-by-side comparison against Mem0, Mastra, Hindsight, Supermemory, or Zep. Those projects publish different metrics on different splits, and placing retrieval recall next to end-to-end QA accuracy is not an honest comparison. See each project's own research page for their published numbers.
Reproducing every result:
git clone https://github.com/MemPalace/mempalace.git
cd mempalace
pip install -e ".[dev]"
# see benchmarks/README.md for dataset download commands
python benchmarks/longmemeval_bench.py /path/to/longmemeval_s_cleaned.json
Knowledge graph
MemPalace includes a temporal entity-relationship graph with validity windows — add, query, invalidate, timeline — backed by local SQLite. Usage and tool reference: mempalaceofficial.com/concepts/knowledge-graph.
MCP server
29 MCP tools cover palace reads/writes, knowledge-graph operations, cross-wing navigation, drawer management, and agent diaries. Installation and the full tool list: mempalaceofficial.com/reference/mcp-tools.
Agents
Each specialist agent gets its own wing and diary in the palace.
Discoverable at runtime via mempalace_list_agents — no bloat in your
system prompt:
mempalaceofficial.com/concepts/agents.
Auto-save hooks
Two Claude Code hooks save periodically and before context compression: mempalaceofficial.com/guide/hooks.
Requirements
- Python 3.9+
- A vector-store backend (ChromaDB by default)
- ~300 MB disk for the default embedding model
No API key is required for the core benchmark path.
Docs
- Getting started → mempalaceofficial.com/guide/getting-started
- CLI reference → mempalaceofficial.com/reference/cli
- Python API → mempalaceofficial.com/reference/python-api
- Full benchmark methodology → benchmarks/BENCHMARKS.md
- Release notes → CHANGELOG.md
- Corrections and public notices → docs/HISTORY.md
Contributing
PRs welcome. See CONTRIBUTING.md.
License
MIT — see LICENSE.