Method Drift›Agent / long-term memory
Nemori
Nemori: Self-Organizing Agent Memory Inspired by Cognitive ScienceAgent / long-term memory · first seen Aug 5, 2025
heavily superseded — a standard baseline that newer methods routinely beat
2 papers critique it · 5 beat it on benchmarks
What papers say
Verbatim critique sentences, each from a paper that cites Nemori as a baseline.
“However, in many such systems, memory access still depends on fixed edge types, manually designed weighting rules, or heuristic traversal procedures.”
— HAGE: Harnessing Agentic Memory via RL-Driven Weighted Graph Evolution“Causal Flatness: Systems like A-MEM and Nemori organize memory based on associative proximity (e.g., semantic links) rather than mechanistic dependency. They can retrieve what happened but struggle to reason about why, as they lack explicit causal modeling.”
— MAGMA: A Multi-Graph based Agentic Memory Architecture for AI Agents
Beaten on benchmarks
Head-to-head results where a newer method reports beating Nemori. Values are copied from the source paper's tables — verify against the cited paper.
- SwiftMem: Fast Agentic Memory via Query-aware Indexing
SwiftMem beats Nemori · BLEU-1 [Temporal Reasoning]
0.569 vs 0.501
- SwiftMem: Fast Agentic Memory via Query-aware Indexing
SwiftMem beats Nemori · BLEU-1 [Overall]
0.467 vs 0.445
- SwiftMem: Fast Agentic Memory via Query-aware Indexing
SwiftMem beats Nemori · Search latency [Overall]
11 vs 835
- HAGE: Harnessing Agentic Memory via RL-Driven Weighted Graph Evolution
HAGE beats Nemori · Overall [gpt-4o-mini]
0.739 vs 0.590
- HAGE: Harnessing Agentic Memory via RL-Driven Weighted Graph Evolution
HAGE beats Nemori · Overall [Qwen2.5-3B]
0.548 vs 0.412
- HAGE: Harnessing Agentic Memory via RL-Driven Weighted Graph Evolution
HAGE beats Nemori · LLM Score [GPT-4o-mini]
0.824 vs 0.624
- HAGE: Harnessing Agentic Memory via RL-Driven Weighted Graph Evolution
HAGE beats Nemori · F1 [GPT-4o-mini]
0.678 vs 0.131
- HAGE: Harnessing Agentic Memory via RL-Driven Weighted Graph Evolution
HAGE beats Nemori · LLM Score [Qwen2.5-3B]
0.527 vs 0.332
- HAGE: Harnessing Agentic Memory via RL-Driven Weighted Graph Evolution
HAGE beats Nemori · F1 [Qwen2.5-3B]
0.429 vs 0.091
- HAGE: Harnessing Agentic Memory via RL-Driven Weighted Graph Evolution
HAGE beats Nemori · Avg. Score [all]
0.739 vs 0.590
- A Simple Yet Strong Baseline for Long-Term Conversational Memory of LLM Agents
event-centric memory beats Nemori · Overall LLM Score [gpt-4o-mini backbone]
0.780 vs 0.744
- A Simple Yet Strong Baseline for Long-Term Conversational Memory of LLM Agents
OursGraph beats Nemori · Overall LLM Score [gpt-4o-mini backbone]
0.780 vs 0.744
Newer alternatives
Recent methods in the same sub-problem, not yet superseded in the knowledge base.
- May 11, 2026
- Mar 16, 2026
- Nov 21, 2025