Categories Companies Research

Loading the SOTA2 catalog…

Explore

Categories Companies Research

Contribute

Join SOTA2 Contact

SOTA2

© 2026 SOTA2

Build in Bellevue, USA

WizWand is now SOTA2 Research

Papers Benchmarks Datasets Tasks

MemoryAgentBench dataset and benchmarks · SOTA2 Research

Research Datasets

MemoryAgentBench

Benchmarks

Task NameDataset NameSOTA MethodSort

Task NameDataset NameSOTA ResultTrendResults

Memory Agent Performance

MemoryAgentBench

83Access Rate

44

Memory-Augmented Agent Evaluation

MemoryAgentBench

28.7Single-Hop Accuracy

16

Long-context memory management

MemoryAgentBench

83Single-Doc Recall

16

End-to-end agent-loop performance

MemoryAgentBench

80Pre-Verify Success Rate

10

Conflict Resolution

MemoryAgentBench conflict-bearing subset official

100Final Accuracy

10

Agentic Memory Retrieval

MemoryAgentBench

82Access Rate

10

Fact Consolidation (Single-Hop)

MemoryAgentBench (MAB) FC-SH 262K

93Accuracy

8

Conflict resolution and memory verification

MemoryAgentBench conflict-bearing slice

100Accuracy

7

Fact Consolidation (Multi-Hop)

MemoryAgentBench (MAB) FC-MH 262K

27Accuracy

5

Fact Consolidation (Single-Hop)

MemoryAgentBench FC-SH Average

0.948Accuracy

4

Fact Consolidation (Single-Hop)

MemoryAgentBench (MAB) FC-SH 64K

95Accuracy

4

Fact Consolidation (Single-Hop)

MemoryAgentBench (MAB) FC-SH 32K

92Accuracy

4

Fact Consolidation (Multi-Hop)

MemoryAgentBench (MAB) FC-MH Average

30.2Accuracy

1

Fact Consolidation (Multi-Hop)

MemoryAgentBench (MAB) FC-MH 64K

33Accuracy

1

Fact Consolidation (Multi-Hop)

MemoryAgentBench FC-MH 32K

27Accuracy

1

Fact Consolidation (Multi-Hop)

MemoryAgentBench (MAB) FC-MH 6K

34Accuracy

1

Popular tasks

Fact Consolidation (Multi-Hop)Fact Consolidation (Single-Hop)Conflict Resolution Long-context memory management Memory Agent Performance

Follow for updates

Join SOTA2 Contact