← Compare Compare · per product
Hydrate vs MemPalace
personal memoryMade verbatim conversation recall mainstream — local ChromaDB/SQLite 'palace', 19 MCP tools, MIT. Per-developer; no decision capture, team canon, or orchestration.
What MemPalace is
| Stack / runtime | Python; ChromaDB + SQLite; 19 MCP tools, auto-save hooks, 5 skills |
| Licence | MIT |
| Storage | ChromaDB + SQLite (local, zero API cost) |
| Hooks | Auto-save hooks |
| Injection | Semantic search via MCP |
| Scope | Single developer; Claude Code, ChatGPT, Cursor via MCP |
| Maturity | Large star count (doc cited ~51.9k; verify) |
| Price | Free (MIT) |
| Version referenced | viral April 2026 launch |
| Source | github.com/MemPalace/mempalace |
Where MemPalace leads
- Verbatim conversation recall (no summarising)
- LongMemEval leadership among free tools (96.6% raw / 100% hybrid, claimed)
- 19 MCP tools + temporal entity-relationship graph with validity windows
- MIT licence
Where Hydrate leads
- Decision-centric memory (what + why), not raw transcript search
- Team canon propagation with attribution
- Cross-vendor live-session memory (not just multi-tool MCP)
- Orchestration; three-layer bootstrap; single dependency-free binary (no ChromaDB)
Benchmarks as published by MemPalace
These are the vendor's own published figures. Hydrate has not
reproduced them and, where the benchmark differs, has no comparable
number, shown as n/a. Listed for transparency, not as a
head-to-head result.
| Benchmark | MemPalace | Hydrate | n | Source |
|---|---|---|---|---|
| LongMemEval (claimed) | 96.6% raw / 100% hybrid | n/a | - | link |
Verification notes
PROVENANCE DISPUTE: internal landscape doc records MemPalace benchmark/provenance claims were 'later disputed'. Treat LongMemEval figures and star counts as claimed, not confirmed. Created by Milla Jovovich & Ben Sigman.
Sources
- Primary repo https://github.com/MemPalace/mempalace
- Homepage https://www.mempalace.tech/