Vibe Coding DiscoverMemoryArena
View on GitHubMemoryArena is a research framework and benchmark for agent memory in interdependent multi-session agentic tasks. It wires pluggable memory backends (long-context, mem0, Letta, Mirix, GraphRAG, MemoRAG, BM25) into task agents and step-based environments for web shopping, travel, search, and formal reasoning.
Use Cases
Benchmarking agent memory across interdependent multi-session tasksComparing memory backends (long-context, mem0, Letta, GraphRAG) on agent tasksEvaluating web shopping agentsEvaluating travel planning agentsEvaluating web search agentsEvaluating formal reasoning agentsWiring memory wrappers around task prompts for LLM agentsRunnable multi-session agentic environment server/client
Built With
- Language
- Python
- Frameworks
- mem0 · mem0-g · Letta · Mirix · ReasoningBank · BM25 · GraphRAG · MemoRAG · text-embedding-rag · OpenAI · Anthropic · Google Gemini · OpenRouter
Tags
agent-memory · benchmark · multi-session · llm-agents · evaluation · memory-systems · long-context · tool-use · environments · research