GitHub radar
Compare AI Memory Agent Memory Leaderboard
agentmemories.ai releases first public leaderboard for long term memory in AI agents, benchmarking retrieval, multi-session dialog, personalization, & temporal reasoning.
The Agent Memory Leaderboard from agentmemories.ai is the first public leaderboard for long term memory in AI agents. We benchmark on 4 key areas of long term memory: retrieval, multi-session dialogue, personalization, and temporal reasoning (i.e. remembering the correct order of events). We have two categories: open source academic models, and production APIs. Our initial launch was in August 2026, but we already have entries from academic researchers and commercial providers. Runs in a Hugging Face Space, no install required!
Why a vibe-coder should care
Great for anyone who wants to build an AI agent, and wants to choose a long term memory system. Helps you make more informed choices instead of having to rely on guesswork. Also useful for seeing how well various LLMs can keep track of the conversation across different inputs.
How to install
Copy this and send it to your agent — Claude Code, Codex, any of them:
Open this space https://huggingface.co/spaces/agent-memory-leaderboard/leaderboard and explain which agent memory systems are benchmarked, what they do, and which one works best for a multi-session project.
Nothing to install — runs in the browser.
Open on Hugging Face▌ More finds