Event ·Multi-source

Mem0 发布 DolphinBench：用准确率、成本、延迟三维度评测智能体记忆

Mem0 发布了智能体记忆基准 DolphinBench。现有评测多把记忆当作问答测验，只报告准确率、precision 和 recall，缺乏对成本和延迟的衡量。DolphinBench 让智能体在长历史上下文之后执行动作并按动作结果打分，把准确率、成本、延迟放在同一张榜上。项目官网为 dolphinbench.ai。

Current read

Mem0 团队做了个新基准 DolphinBench，专测智能体记忆：不看答题对错，看长对话后动作做没做对，还把成本和延迟一起排名，选记忆系统前可以先看看。

3 sources73° AI heatLast updated 2026/9/22 16:41:25

Evidence

3 sources

Primary sourcemem0

2026/9/22 16:41:25View source

Related source 1Taranjeet

2026/9/22 18:06:41View source

Related source 2arXiv cs.AI

2026/9/21 17:54:35View source

Conflict check

Dedup data only shows that these sources discuss the same event, not that they agree. There is no structured supporting or refuting evidence yet, so no conflict is inferred automatically.

Open the story