Production-ready memory infrastructure that takes 60 seconds to integrate. No complex setup, no vendor lock-in, no surprises.
Lightning-fast memory retrieval with p95 latency under 200ms. No more waiting 20+ seconds for context.
Combine semantic and keyword search for accurate results. Get the best of both worlds automatically.
Manage memory for unlimited agents. Each agent gets isolated, searchable memory with zero cross-contamination.
Works with LangChain, LlamaIndex, or direct API calls. Four simple methods: add, search, update, delete.
Track memory usage, search performance, and agent activity. Real-time insights into your AI systems.
From prototype to production without code changes. Infrastructure scales automatically with your usage.
Low-latency access worldwide. Your memories are replicated across multiple regions for reliability.
Native support for React Native. Build mobile AI agents with persistent memory in minutes.
Deploy BaseGrid on your own infrastructure with Docker. Keep your data completely within your VPC.
Auto-categorize memories and expand search queries using Google Gemini for better recall.
Four methods to manage memory. That's it. No complex configuration, no learning curve.
from basegrid import BaseGrid
# Initialize
grid = BaseGrid(api_key="your_api_key")
# Add memory
grid.memory.add(
agent_id="agent_1",
content="User prefers dark mode",
metadata={"category": "preferences"}
)
# Search memory
results = grid.memory.search(
agent_id="agent_1",
query="What does the user prefer?",
limit=5
)
# That's it! 🚀Start building in under 60 seconds. Free tier includes 3 agents and 100K API calls.