A Markdown wiki outscored every AI agent memory product we benchmarked

A benchmark by an independent testing group found that a Markdown wiki outperformed every AI agent memory product tested, with Claude Code's built-in memory scoring 67.7% accuracy and costing $426.90 per 1,000 successful answers. The Agentic Memory Index v0.1, published in August 2026, measures accuracy, cost, and speed for each tool using the same agent model.

See how much smarter your AI could be by adding tools We independently measure the tools AI agents use and publish what each one actually adds. Agentic Memory Indexi Methodology /methodology memory Agentic Memory Index v0.1 · Aug 2026 Highlights Accuracy, cost, and speed for every tool, measured the same way. Agentic Memory Indexi The overall score: how often each tool finds the right answer. Claude Code built-in memory 67.7 Costi USD per 1,000 successful answers Claude Code built-in memory $426.90 Answer speedi How long an answer takes end to end, same agent model on every tool.