Intelligence,
Engineered.
Add persistent memory to any LLM in 3 lines of code.
Indexless Memory
Developers never touch a vector index. ICE handles embedding, storage, and retrieval entirely under the hood.
Zero-Conf Local Mode
Boots instantly with an embedded DuckDB file (~/.ice/ice_memory.db). No Docker, Postgres, or daemons required.
Drop-In SDK Middleware
Wraps standard AsyncOpenAI or Anthropic Python clients in exactly 3 lines of code. No new API paradigms to learn.
One import.
Persistent Memory.
ICE wraps standard LLM client libraries to automatically handle embedding, database storage, and context retrieval. No database setup, chunking logic, or index management needed.
Transparent LLM Client Wrapper
Zero-Conf Local DuckDB Storage
Multi-Platform Native Core C-Extension
import ice
import asyncio
import os
# 1. Set your signed license key
os.environ["ICE_LICENSE_JWT"] = "your_signed_jwt_here"
async def main():
# 2. Ingest documents into memory
await ice.ingest("s3://data-lake/architecture.pdf")
# 3. Query with automatic memory retrieval
response = await ice.query(
prompt="Analyze this.",
model="gpt-4o"
)
print(response)
asyncio.run(main())
Core Capabilities
The architectural advantages of Persistent Context Engineering
Velocity. Pure.
Low latency context retrieval and management. Native intelligence at speed.
Identity. Persistent.
Transform stateless interactions into evolving, stateful agent identities.
Reliability. Absolute.
Secure, isolated, and scalable memory architecture for the next decade of B2B.
Built for each other.
ICE
Indexless Context Engine. Perfect memory and context ingestion for every model.
Cortex
Hardened Runtime. A stable, high-performance engine for deploying complex agent ecosystems.
GoingMerry
Fastest & Efficient Local AI Runtime. Run any model privately on your machine.