You brainstorm in ChatGPT, debug in Cursor, try a new coding agent and re-explain everything from scratch. With every new AI tool, the cost of context switching grows.
Just 2 months after launch, weโve built an open source personal memory layer that accurately recalls context for each individual and can be used across multiple apps like ChatGPT, claude, cursor, gemini, claude code etc.
CORE memory scored 88.24% ๐๐๐๐ฎ๐ซ๐๐๐ฒ ๐ข๐ง ๐ฆ๐๐ฆ๐จ๐ซ๐ฒ ๐ซ๐๐๐๐ฅ๐ฅ in LoCoMo benchmark.
LoCoMo tests how well AI systems remember and reason across long conversations (300+ turns). Think of it as the SAT for AI memory - it evaluates whether systems can maintain context, resolve contradictions, and surface relevant information as conversations evolve over time.
Results:
- Single-hop: 91% (Simple recall โWhatโs your favorite framework?โ)
- Multi-hop: 85% (Connecting facts โWho else uses React on your team?โ)
- Temporal: 88% (Tracking changes โWhen did you switch to Next.js?โ)
- Open-domain: 71% (General world knowledge)
- Overall: 88%
Where you can use CORE
Every AI tool you already use: ChatGPT, Claude, Cursor, Gemini, Claude Code - via MCP integration.
Key workflows:
Debugging: Remember your codebase patterns across sessions
Code reviews: Track team standards and architectural decisions
Research: Connect new concepts to what you've learned
Multi-tool: Brainstorm in ChatGPT โ code in Cursor โ deploy with Claude Code
No more re-explaining context. CORE remembers not just what you built, but why.
CORE is open-source
Weโre building CORE in public and itโs fully open source. If you share our vision of memory that belongs to individuals, not platforms:
๐ Try CORE today:
- Quick start: Sign up at core.heysol.ai
- Self-host: Run locally with docker compose up
Own your memory. Power your agents.
Top comments (0)