Getting Started
July 30, 2026 · View on GitHub
Requirements
- Python 3.10 or later
- (Optional) Hermes Agent Framework for plugin integration
Installation
From PyPI (recommended)
pip install mnemosyne-memory
With all optional features (dense retrieval via fastembed + local LLM consolidation):
pip install mnemosyne-memory[all]
Ubuntu 24.04 / Debian 12: If pip install fails with externally-managed-environment, use a virtual environment:
python3 -m venv .venv
source .venv/bin/activate
pip install mnemosyne-memory[all]
From Source (for contributors)
git clone https://github.com/AxDSan/mnemosyne.git
cd mnemosyne
pip install -e ".[all,dev]"
Hermes Provider
pip install mnemosyne-hermes
hermes config set memory.provider mnemosyne
hermes gateway restart
No deploy script needed. The pip package handles everything.
Your First Memory
from mnemosyne import remember, recall
# Store a fact
remember("User prefers dark mode interfaces", importance=0.9, source="preference")
# Store a global fact (visible in every session)
remember("User email is alice@example.com", importance=0.95, source="profile", scope="global")
# Store temporary data with expiry
remember("API key: sk-abc123", importance=0.8, source="credential", valid_until="2026-12-31T00:00:00")
# Search memories
results = recall("interface preferences", top_k=3)
for r in results:
print(r["content"])
Verify Installation
from mnemosyne import get_stats
print(get_stats())
Next Steps
- Architecture — understand how BEAM tiers work
- API Reference — full Python API documentation
- Configuration — tune performance and storage
- Hermes Integration — use as a Hermes memory backend