List your company
ResearchFormerly WizWand

Long-horizon conversation utility evaluation on LongMemEval

77.8AccuracyMemOS
42.117651.381360.64569.9087May 27, 2026
Updated 5mo ago

Evaluation results

MethodLinks
MemOSBase LLM=GPT-4o-mini, Judge LLM=GPT-4o-mini, # Avg. Token=15892026.0577.8
MEMGUARDBase LLM=GPT-4.1-mini, Judge LLM=GPT-4.1, # Avg. Token=16052026.0573.5
MemobaseBase LLM=GPT-4o-mini, Judge LLM=GPT-4o-mini, # Avg. Token=21022026.0572.4
MEMGUARD (2 hop)Base LLM=GPT-4.1-mini, Judge LLM=GPT-4.1, # Avg. Token=15402026.0572.25
Mem0Base LLM=GPT-4o-mini, Judge LLM=GPT-4o-mini, # Avg. Token=11722026.0566.4
ZepBase LLM=GPT-4o-mini, Judge LLM=GPT-4o-mini, # Avg. Token=27012026.0563.8
A-MemBase LLM=GPT-4.1-mini, Judge LLM=GPT-4.1, # Avg. Token=72442026.0562.25
MEMGUARDBase LLM=GPT-4o-mini, Judge LLM=GPT-4o-mini, # Avg. Token=12972026.0560
MEMGUARD (2 hop)Base LLM=GPT-4o-mini, Judge LLM=GPT-4o-mini, # Avg. Token=12502026.0560
SupermemoryBase LLM=GPT-4o-mini, Judge LLM=GPT-4o-mini, # Avg. Token=5002026.0558.4
A-MemBase LLM=GPT-4o-mini, Judge LLM=GPT-4o-mini, # Avg. Token=72442026.0547
MIRIXBase LLM=GPT-4o-mini, Judge LLM=GPT-4o-mini2026.0543.49