# How should we evaluate image memory in AI agents?

> Source: <https://twitter.com/AgentMemoryL/status/2123418533950591252>
> Published: 2026-11-20 06:24:56+00:00

**Recommendation:** Benchmark agent memory on separate text, code and multimodal tracks rather than one aggregate score, as AML Cycle 2 did across 402 task submissions.
More in the playbook "Evals and testing for LLM apps": <https://wpnews.pro/playbooks/evals-testing>

🏆 AML Cycle 2 results are live! 💥
• 1M+ cross-platform impressions 
• 10M+ cumulative website clicks 
• 402 task submissions 
• Three tracks: text, code and multimodal memory 
This cycle brought together teams from across the global AI community, including teams from Alibaba Cloud DMS, Shopee, OPPO, KT and OceanBase—alongside university researchers, memory startups and independent developers.
Explore the full results:
