Give Your AI a Memory
That Never Leaves Your Machine
Offline semantic memory engine. Single Rust binary. ~45ms recall. Zero API keys.
Production v0.7.2 Single Rust Binary Zero API Keys
Built for Speed & Privacy
Every feature designed around one principle: your data stays local, recalls are instant.
Semantic Search
EmbeddingGemma 768d, hybrid FTS5+vector retrieval
~45ms Recall
SQLite + usearch, zero network round-trip — recalls happen locally in milliseconds
Single Binary
curl install, no Docker, no Python, no Node
MCP Native
27 tools, works with Claude Code, Cursor, Copilot
Local-First
Your data never leaves ~/.uteke/
Auto-Extraction
Session transcripts to structured memories automatically
Quickstart
Three commands. 30 seconds. You're done.
# Install
curl -fsSL codecora.dev/install | sh
# Remember
uteke remember "Deploy v2.1 to staging" --tags deploy
# Recall (~45ms)
uteke recall "when did we deploy?"Architecture
Zero dependencies. Zero network. Pure local.
uteke single binary
SQLite structured storage
usearch vector index
EmbeddingGemma 768d ONNX
uteke-serve HTTP API :8767
uteke-mcp MCP server · 27 tools
Uteke vs Everyone Else
Sourced from the Uteke README. Same privacy thesis, zero of the cloud baggage.
| Metric | Uteke | Mnemosyne | Mem0 | AgentMemory | Letta | Zep | Engram |
|---|---|---|---|---|---|---|---|
| Language | Rust (single binary) | Python (pip) | Python | TypeScript | Python | Python | Go (single binary) |
| Setup | One binary (curl | sh) | pip install + venv | pip + Docker + Qdrant | npm + Docker (iii-engine) | pip + Docker + Postgres | pip + Docker + Neo4j | One binary |
| API keys | None | For remote embeddings | OpenAI/LLM | LLM key | LLM key | LLM key | None |
| Works offline | Fully | Optional | Cloud embedding | Needs LLM | Needs LLM | Needs LLM + vector DB | Fully |
| Search | Hybrid (Vector + FTS5 + RRF) | sqlite-vec + FTS5 | Vector + Graph | Vector + Graph | Vector | Temporal Graph | FTS5 only |
| Recall speed | ~45ms | ~50ms+ | Network round-trip | Network round-trip | Network round-trip | Network round-trip | ~Fast (local) |
| Multi-agent | Rooms (built-in) | Shared API | — | — | — | — | — |
| Time-travel | Native point-in-time | Temporal triples | — | — | — | — | — |
| MCP server | JSON-RPC + HTTP | stdio + SSE | — | — | — | — | — |
| Your data | Never leaves machine | Local-first | Sent to LLM cloud | Sent to LLM cloud | Sent to LLM cloud | Sent to LLM cloud | Local |
| License | Apache 2.0 | MIT | Apache 2.0 | Apache 2.0 | Apache 2.0 | Apache 2.0 | Apache 2.0 |
Source: Uteke README. Recall speed reflects local single-machine recall; cloud tools add network latency.