Agent Memory Leaderboard｜智能体记忆系统公开评测
An open Add/Search evaluation framework for agent memory
- Provides an open Add/Search evaluation framework to benchmark agent memory across text, code, and multimodal tracks.
- Uses a unified contract to test memory retrieval and addition, enabling consistent comparison between systems.
- Makes results publicly available, fostering transparency and accelerating research in agent memory capabilities.
文本、代码、多模态三赛道，以统一 Add / Search 契约检验 Agent 记忆能力。
Visit Agent Memory Leaderboard｜智能体记忆系统公开评测
More in Agents
- Agent-Reach — Give your AI agent eyes to see the entire internet. Read & search Twitter, Reddit,…
- WeKnora — Open-source LLM knowledge platform: turn raw documents into a queryable RAG, an…
- OpenMontage — World's first open-source, agentic video production system. 12 production pipelines,…
- hermes-agent — The agent that grows with you
- oh-my-pi — ⌥ Coding agent with the IDE wired in. Built by Stencil Labs.
- ppt-master — AI turns documents or topics into real, native PowerPoint decks—with native shapes,…
- TradingAgents — TradingAgents: Multi-Agents LLM Financial Trading Framework
- OpenHands — 🙌 OpenHands: AI-Driven Development
Opening Liz…
