List your company
ResearchFormerly WizWand

Long-term memory question answering on LoCoMo 0.6.0

94.48AccuracyEverMemOS
60.10869.031577.95586.8785May 6, 2026
Updated 6mo ago

Evaluation results

MethodLinks
EverMemOSAnswer model=gpt-4.1-mini, Judge model=gpt-4o-mini, Judge runs / question=3 (majority), Answer temperature=0, Judge temperature=0, Max answer tokens=200, Retrieval top-k=100, Retrieval infra=proprietary cloud infrastructure (4-billion-parameter GPU-hosted embedder)2026.0594.4893.231,4550.001——
TM ProAnswer model=gpt-4.1-mini, Judge model=gpt-4o-mini, Judge runs / question=3 (majority), Answer temperature=0, Judge temperature=0, Max answer tokens=200, Retrieval top-k=100, 3-run means=true2026.059391.61,4320.00134.090.043
TM BaseAnswer model=gpt-4.1-mini, Judge model=gpt-4o-mini, Judge runs / question=3 (majority), Answer temperature=0, Judge temperature=0, Max answer tokens=200, Retrieval top-k=100, 3-run means=true2026.0592.0190.541,4170.00111.120.29
TM EdgeAnswer model=gpt-4.1-mini, Judge model=gpt-4o-mini, Judge runs / question=3 (majority), Answer temperature=0, Judge temperature=0, Max answer tokens=200, Retrieval top-k=100, 3-run means=true2026.0589.6588.051,3810.0016.540.011
RAG (ChromaDB)Answer model=gpt-4.1-mini, Judge model=gpt-4o-mini, Judge runs / question=3 (majority), Answer temperature=0, Judge temperature=0, Max answer tokens=200, Retrieval top-k=1002026.0586.1784.351,3270.001111.566.7
EngramAnswer model=gpt-4.1-mini, Judge model=gpt-4o-mini, Judge runs / question=3 (majority), Answer temperature=0, Judge temperature=0, Max answer tokens=200, Retrieval top-k=1002026.0584.5582.651,3020.00112.280.13
BM25Answer model=gpt-4.1-mini, Judge model=gpt-4o-mini, Judge runs / question=3 (majority), Answer temperature=0, Judge temperature=0, Max answer tokens=200, Retrieval top-k=1002026.0580.4578.41,2390.001117.393
Zep (published)note=using their own evaluation methodology, not the paper harness2026.0571—————
SupermemoryAnswer model=gpt-4.1-mini, Judge model=gpt-4o-mini, Judge runs / question=3 (majority), Answer temperature=0, Judge temperature=0, Max answer tokens=200, Retrieval top-k=1002026.0565.3962.981,0070.001998.0910
Mem0Answer model=gpt-4.1-mini, Judge model=gpt-4o-mini, Judge runs / question=3 (majority), Answer temperature=0, Judge temperature=0, Max answer tokens=200, Retrieval top-k=1002026.0561.4358.979460.00316.260.012