hindsight vs MemPalace
Two of the top memory, side by side: score, setup, license, activity and what each review found.
hindsight
Agent memory server with retain, recall and reflect operations
MemPalace
Local verbatim memory for coding agents on ChromaDB with 45 MCP tools
| What we compare | hindsight | MemPalace |
|---|---|---|
| Score parts, out of 100 | ||
| Adoption | 81, widely used | 87, widely used |
| Freshness | 100, active | 100, active |
| Maintenance | 93, healthy | 84, healthy |
| Easy to run | 50, easy | 50, easy |
| Agent-ready | 85, ready | 70, partly |
| Facts from GitHub and the README | ||
| Stars | 48.3k | 59.5k |
| License | MIT (permissive) | MIT (permissive) |
| Last commit | Oct 2026 | Oct 2026 |
| Last release | Oct 2026 | Sep 2026 |
| Language | Python | Python |
| Docker | Yes | Yes |
| GPU | Not stated | Not needed |
| arm64 or Apple Silicon | Mentioned | Mentioned |
hindsight
Hindsight stores agent memories in banks and extracts facts, entities and timestamps from retained text using an LLM. Recall runs semantic, BM25, graph and temporal retrieval in parallel, then reranks the merged results. Background jobs consolidate facts into observations and mental models. It exposes a REST API, Python, Node.js and Go clients, a CLI, and a per-bank MCP endpoint.
Who it is for: Developers adding long-term memory to LLM agents and coding assistants
Strengths
- Four parallel retrieval strategies merged with reciprocal rank fusion and cross-encoder reranking
- Works with 25+ LLM providers, including local ollama, lmstudio and llamacpp
- Built-in MCP endpoint per bank, plus 60+ listed integrations
- Embedded mode runs in-process via pip with a bundled pg0 database
Weaknesses
- Every retain call requires an LLM, adding cost and latency
- Accuracy claims are the vendor's own benchmarks; independent reproduction is partial
- Managed Cloud and Enterprise tiers exist; feature differences from self-hosted are unclear
- Minimum RAM and GPU needs are not stated in the README
- Docker
- Needs LLM provider (hosted or local), PostgreSQL (embedded pg0 by default), Oracle AI Database (optional)
- Models: openai, anthropic, gemini, groq, bedrock
- port 9999
MemPalace
MemPalace stores conversation history verbatim, never summarised, and retrieves it by scoped semantic search over a palace of wings (people, projects), rooms (topics) and drawers. It runs locally with Python 3.9+ and ChromaDB by default, exposes 45 MCP tools plus a CLI, mines Claude Code, Codex and Cursor transcripts via hooks, and needs no API key: 96.6% R@5 on LongMemEval without an LLM.
Who it is for: Coding-agent users who want local, inspectable, verbatim session memory
Strengths
- Verbatim storage; nothing is summarised or paraphrased
- 96.6% R@5 on LongMemEval with no LLM or API key; results reproducible from the repo
- Pluggable backends: ChromaDB, sqlite, Rust-native, Milvus, Qdrant, pgvector
- Multi-arch Docker image; auto-save hooks for Claude Code, Codex and Cursor
Weaknesses
- First run downloads an 80 to 300 MB embedding model; Docker needs network then
- No native Android/Termux; GPU image is x86_64-only and unpublished
- Docker image runs as uid 1000, so bind mounts must be readable by that uid
- README warns about impostor domains distributing malware
- no GPU
- Docker + Compose
- Models: local embeddings (MiniLM, EmbeddingGemma), OpenAI-compatible embedding endpoints, Ollama