Skip to main content

Configuration

Environment Variables​

LanceDB Storage​

# Path to LanceDB storage (default: .suzent/data/memory)
# Can be relative or absolute path
LANCEDB_URI=.suzent/data/memory

Embedding​

EMBEDDING_MODEL=text-embedding-3-large  # or text-embedding-3-small
EMBEDDING_DIMENSION=3072                # Auto-detected from CONFIG if omitted

Memory System​

MEMORY_ENABLED=true                    # Enable/disable memory system
EXTRACTION_MODEL=gpt-4o-mini          # LLM for fact extraction (optional)
USER_ID=default-user                   # Default user identifier

API Keys​

OPENAI_API_KEY=sk-xxx

Manager Initialization​

from suzent.memory import MemoryManager, LanceDBMemoryStore

# Initialize store
store = LanceDBMemoryStore(
    uri=".suzent/data/memory",
    embedding_dim=3072
)
await store.connect()

manager = MemoryManager(
    store=store,
    embedding_model="text-embedding-3-large",
    embedding_dimension=3072,  # Optional
    llm_for_extraction="gpt-4o-mini"  # Optional
)

System Constants​

Defined in manager.py:

DEFAULT_MEMORY_RETRIEVAL_LIMIT = 5
DEFAULT_MEMORY_SEARCH_LIMIT = 10
IMPORTANT_MEMORY_THRESHOLD = 0.7
DEDUPLICATION_SIMILARITY_THRESHOLD = 0.85
DEFAULT_IMPORTANCE = 0.5

Tuning​

More Aggressive Storage​

DEDUPLICATION_SIMILARITY_THRESHOLD = 0.75
DEFAULT_IMPORTANCE = 0.6

Cleaner Memory​

DEDUPLICATION_SIMILARITY_THRESHOLD = 0.90
DEFAULT_IMPORTANCE = 0.3

Hybrid Search Weights​

results = await store.hybrid_search(
    ...,
    semantic_weight=0.7,      # Vector similarity
    fts_weight=0.3,           # Full-text
    importance_boost=0.2,     # Importance
    recency_boost=0.1         # Recency
)

Embedding Models​

ModelDimensionCost/1M tokensUse Case
text-embedding-3-large3072$0.13Production
text-embedding-3-small1536$0.02Development
text-embedding-ada-0021536$0.10Legacy

Storage Management​

Check Storage Size​

import os
from pathlib import Path

db_path = Path(".suzent/data/memory")
if db_path.exists():
    total_size = sum(f.stat().st_size for f in db_path.rglob('*') if f.is_file())
    print(f"Memory storage: {total_size / (1024**2):.2f} MB")

Clear All Memories​

# Delete all memories for a user
await store.delete_all_memories(user_id="user-123")

Debug Logging​

import logging
logging.basicConfig(level=logging.DEBUG)