Spaces:
Running
Running
Add Agent Memory Leaderboard organization card
Browse files
README.md
CHANGED
|
@@ -1,10 +1,60 @@
|
|
| 1 |
---
|
| 2 |
-
title:
|
| 3 |
-
emoji:
|
| 4 |
colorFrom: blue
|
| 5 |
-
colorTo:
|
| 6 |
sdk: static
|
| 7 |
pinned: false
|
|
|
|
| 8 |
---
|
| 9 |
|
| 10 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
---
|
| 2 |
+
title: Agent Memory Leaderboard
|
| 3 |
+
emoji: 🧠
|
| 4 |
colorFrom: blue
|
| 5 |
+
colorTo: green
|
| 6 |
sdk: static
|
| 7 |
pinned: false
|
| 8 |
+
short_description: A unified evaluation platform for memory systems.
|
| 9 |
---
|
| 10 |
|
| 11 |
+
# Agent Memory Leaderboard · 记忆之巅
|
| 12 |
+
|
| 13 |
+
**A unified evaluation platform for long-term memory systems and memory-enabled agents.**
|
| 14 |
+
|
| 15 |
+
Agent Memory Leaderboard (a.k.a. 记忆之巅) evaluates how effectively memory systems store, retrieve, and use information across long conversations, persistent user contexts, and memory-intensive agent tasks.
|
| 16 |
+
|
| 17 |
+
> Official evaluation and submission are conducted exclusively through the [Agent Memory Leaderboard website](https://agentmemoryleaderboard.ai/). This Hugging Face organization hosts public leaderboard releases, evaluation documentation, and community updates.
|
| 18 |
+
|
| 19 |
+
## What We Evaluate
|
| 20 |
+
|
| 21 |
+
- Long-term memory storage and retrieval
|
| 22 |
+
- Long-context and multi-session understanding
|
| 23 |
+
- Personalized and user-specific memory
|
| 24 |
+
- Temporal and event-ordering reasoning
|
| 25 |
+
|
| 26 |
+
## Evaluation Tracks
|
| 27 |
+
|
| 28 |
+
| Track | Intended for |
|
| 29 |
+
|---|---|
|
| 30 |
+
| **Academic Methods** | Reproducible research systems, open-source methods, and academic implementations |
|
| 31 |
+
| **Industry Systems** | Production APIs, hosted services, commercial systems, and closed-source products |
|
| 32 |
+
|
| 33 |
+
All systems are evaluated under a versioned evaluation contract with fixed datasets, prompts, answer models, judge configurations, and pipeline hashes.
|
| 34 |
+
|
| 35 |
+
## First Public Release
|
| 36 |
+
|
| 37 |
+
The inaugural public leaderboard is scheduled for **mid-August 2026**.
|
| 38 |
+
|
| 39 |
+
Results published on Hugging Face are official release snapshots. The live and authoritative leaderboard remains on the Agent Memory Leaderboard website.
|
| 40 |
+
|
| 41 |
+
## Links
|
| 42 |
+
|
| 43 |
+
- 🌐 [Official Website](https://agentmemoryleaderboard.ai/)
|
| 44 |
+
- 🚀 [Evaluation Portal](https://agentmemoryleaderboard.ai/evaluation)
|
| 45 |
+
- 🏆 [Public Leaderboard](https://agentmemoryleaderboard.ai/leaderboard/textual)
|
| 46 |
+
- 📖 [Evaluation Documentation](https://agentmemoryleaderboard.ai/docs)
|
| 47 |
+
- 💻 [GitHub Repository](https://github.com/tjuniko/agent-memory-leaderboard)
|
| 48 |
+
- 𝕏 [Follow Agent Memory Leaderboard](https://x.com/AgentMemoryL)
|
| 49 |
+
|
| 50 |
+
## On Hugging Face
|
| 51 |
+
|
| 52 |
+
This organization will publish:
|
| 53 |
+
|
| 54 |
+
- Versioned leaderboard result datasets
|
| 55 |
+
- Public leaderboard Spaces
|
| 56 |
+
- Benchmark and methodology documentation
|
| 57 |
+
- Baselines and reproducibility materials
|
| 58 |
+
- Release announcements and technical analyses
|
| 59 |
+
|
| 60 |
+
Follow this organization for the inaugural leaderboard release and future evaluation cycles.
|