Agent Memory Leaderboard (@AgentMemoryL) header
Agent Memory Leaderboard (@AgentMemoryL) profile picture

Agent Memory Leaderboard Official. First open championship for agent memory systems. Neutral · Reproducible · Fair. mailto: [email protected]

Joined July 2026
How do we evaluate long-term memory in AI agents? Storing information is one thing. Remembering the right information, adapting to updates, resolving conflicts, and applying past experience to new tasks is another. The 2nd Agent Memory Leaderboard (AML) aims to provide a systematic evaluation of these capabilities. This edition has received submissions from 312 memory systems across three evaluation tracks: 🧠 Text Memory 6,000+ evaluation instances, including a new Streaming Memory evaluation that tests how systems handle continuously evolving information. 💻 Code Memory 300 evaluation units across 150 base tasks, testing the use of relevant and noisy memory in bug fixing and feature development. 🌐 Multimodal Memory Nearly 5,000 evaluation instances, focusing on image-text memory, cross-modal retrieval, and evidence use. The full evaluation and result verification process has been completed. The full rankings, track-level performance analysis, and selected technical approaches will be published on November 19, 2026. We hope AML can help make long-term agent memory capabilities more measurable and comparable. Explore AML: [agentmemories.ai/home]
3
4
110
A customer explains a business rule once. Will an AI agent still remember it when making a decision several rounds later? FDE-Bench highlights a subtle warehouse challenge: two batches of the same product may still count as mixed goods under the warehouse's rules. An agent that groups only by product ID could get the entire plan wrong. Remembering a rule isn't enough. The agent must retrieve and apply it when the decision actually depends on it. This is exactly the kind of capability we evaluate in the Textual Memory track of Agent Memory Leaderboard (AML). AML Cycle 2 results are expected in mid-November. Stay tuned to see how participating memory systems perform! Research: arxiv.org/html/2610.14539v3#… AML: github.com/AML-memory/agent-…
6
81
Coding agents are getting faster. But speed is only one part of maintaining a real project. The harder question: after fixing a bug today, can an agent use that experience when a related task appears next week? Zed’s Agent Metrics makes responsiveness visible. Project memory asks what happens over time.
1
2
84
Finding the right memory doesn't guarantee the right answer. An October 8 analysis by Engram tested 6 memory systems on 1,536 LoCoMo questions, evaluating retrieval and answer accuracy separately. The takeaway: retrieval success ≠ answer correctness. Report: engram.tools/en/insights/ai-…
4
4
169
⏰ 2 days left to complete your AML Cycle 2 submission. All materials must be submitted and after October 31, 23:59 (UTC+8), the platform will no longer accept newly started Full Evaluation jobs. From November 1 to November 4, only already-started jobs may resume from checkpoints. The evaluation cycle closes on November 4, 23:59 (UTC+8). Tracks: • Textual Memory • Coding Memory • Multimodal Memory Teams must also complete real-name and team information registration on the AML platform by November 4. This information will be used for official leaderboard attribution, result notifications, and award eligibility review. Commercial entries will also be displayed on the AML website with their product links. Submit here: agentmemoryleaderboard.ai/co…
3
1
171
Why can the same Agent Memory system get different scores across reports? Redis reran memory systems on LongMemEval under a common judge and found gaps from previously published numbers. Models, prompts, retrieval settings, and scoring can all matter. Report: redis.io/resources/agent-mem…
🤖 Made with AI
1
1
8
210
11 days left to join AML Cycle 2. The deadline for complete submissions is October 31, 2026, 23:59 UTC+8. We evaluate Agent Memory across: • Textual Memory • Coding Memory • Multimodal Memory Participants bring their own Add / Search systems. AML standardizes downstream Answer / Eval, review, and public reporting. Especially interested in systems that can handle more than recall: stale information, conflicting evidence, noisy repository history, and long-running tasks. Learn more and apply: agentmemoryleaderboard.ai/
2
4
129
Agent Memory Challenge 2026 Cycle 2 is now open. Long-term memory is not just about storing more history. It is about retrieving the right evidence, recognizing what has changed, and avoiding stale context when an agent needs to act. Three tracks: Textual · Coding · Multimodal Open-source Methods · Commercial Products Over USD 22,000 prize pool for eligible open-source teams. A shared Add/Search interface. Standardized Answer/Eval. Public, comparable results. Join: agentmemoryleaderboard.ai/ev…
19
7
10
36
57,076
What should Agent Memory be judged on? Not whether a system can write a convincing final answer—but whether it can reliably retrieve the right evidence from long-running experience. In 2 days, Cycle 2 of the Agent Memory Challenge opens. One shared Add/Search interface. One standardized Answer/Eval pipeline. Textual, Coding, and Multimodal Memory. Measure memory. Compare what matters. September 20, 00:00 UTC+8 agentmemoryleaderboard.ai/
🤖 Made with AI
3
3
3
9
339
The three tracks have a total prize pool of ¥150,000 (~$21K USD). The prize pool is for the Open-Source Methods rankings. Registration opens: September 20, 2026 Evaluation & Key application: agentmemoryleaderboard.ai/ev… Rules: agentmemoryleaderboard.ai/ru… API Guide: agentmemoryleaderboard.ai/ap… GitHub: github.com/AML-memory/agent-… Think your memory system can compete? Build it. Submit it. See where it stands.
15
Season 2 covers three tracks: Text Long-term dialogue, factual recall, multi-hop reasoning, personalization, and memory safety. Code Debugging memory, development experience, and project context for software engineering agents. Multimodal Memory and cross-modal retrieval across image, audio, video, and text. Each track has separate rankings for Open-Source Methods and Commercial Products.
1
47
Agent Memory Challenge — Season 2 is coming. ~$21K USD prize pool. 3 tracks. 1 unified evaluation platform. Hosted by CSIG, the 2nd Agent Memory Challenge is open to researchers, developers, open-source teams, and individual participants worldwide. Tracks: Text Code Multimodal Open-source methods and commercial products will be evaluated separately under a unified Add / Search protocol. Registration opens: September 20, 2026 Build your memory system. Put it to the test. See where it stands. Register & apply for an Evaluation Key: agentmemoryleaderboard.ai/ev…
1
4
199
Nice to see Agent Memory Leaderboard showing up on the radar 📈 AML was recently ranked #1 among trending Hugging Face Spaces over the past 7 days by Indie Signals. Thanks to everyone exploring and contributing to the benchmark. More to come.
2
91
The first AML results are live, and the community discussion has raised many important questions around evaluating AI memory systems. Beyond rankings, we’re interested in deeper questions: • Can improvements be attributed to the memory system itself? • Are evaluations reproducible? • How should we measure memory beyond retrieval? Thanks to everyone sharing feedback and pushing the conversation forward.
43