You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Copy file name to clipboardExpand all lines: README.md
+1-1Lines changed: 1 addition & 1 deletion
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -17,7 +17,7 @@ memweave is a zero-infrastructure, async-first Python library that gives AI agen
17
17
18
18
## 📊 Benchmark — LongMemEval-S
19
19
20
-
Evaluated on [LongMemEval-S](https://huggingface.co/datasets/xiaowu0162/longmemeval-cleaned) — a 500-question benchmark covering multi-session memory, temporal reasoning, knowledge updates, and user preferences. Primary metric: **Recall@5**.
20
+
Evaluated on [LongMemEval-S](https://huggingface.co/datasets/xiaowu0162/longmemeval-cleaned) — a 500-question benchmark covering multi-session memory, temporal reasoning, knowledge updates, and user preferences. Primary metric: retrieval recall i.e. **Recall@k** (correct session in the top-k results).
21
21
22
22
**Embedding model used:**`all-MiniLM-L6-v2` via Ollama (local), same as mempalace. **No LLM, no API key, no cloud at any stage.**
0 commit comments