List your company
ResearchFormerly WizWand

DB Task on Lifelong Agent Bench Task

86Success RateQWEN3.5-397B-A17B
61.5667.90574.2580.595Jun 16, 2026
Updated 4mo ago

Evaluation results

MethodLinks
QWEN3.5-397B-A17BBackbone LLM=QWEN3.5-397B-A17B2026.0686——
OPD-Evolver-9BBackbone LLM=QWEN3.5-9B2026.0684.5——
EvolveRBackbone LLM=QWEN3.5-9B2026.0682.5——
MempBackbone LLM=QWEN3.5-9B2026.0682——
STEP-3.5-FLASH (196B)Backbone LLM=STEP-3.5-FLASH (196B)2026.0681——
MemEvolveBackbone LLM=QWEN3.5-9B2026.0681——
ReasoningBankBackbone LLM=QWEN3.5-9B2026.0680.5——
CheatsheetBackbone LLM=QWEN3.5-9B2026.0679.5——
No MemoryBackbone LLM=QWEN3.5-9B2026.0675.5——
OPD-Evolver-4BBackbone LLM=QWEN3-4B2026.0674——
ExpeLBackbone LLM=QWEN3.5-9B2026.0674——
MempBackbone LLM=QWEN3-4B2026.0673——
AWMBackbone LLM=QWEN3.5-9B2026.0672.5——
MemEvolveBackbone LLM=QWEN3-4B2026.0672——
EvolveRBackbone LLM=QWEN3-4B2026.0671.5——
ReasoningBankBackbone LLM=QWEN3-4B2026.0670——
CheatsheetBackbone LLM=QWEN3-4B2026.0669——
No MemoryBackbone LLM=QWEN3-4B2026.0665——
AWMBackbone LLM=QWEN3-4B2026.0663——
ExpeLBackbone LLM=QWEN3-4B2026.0662.5——
Mem0Model=GPT-4o-mini2026.01—9292.6
MemPModel=GPT-4o-mini2026.01—9696.6
MemRLModel=GPT-4o-mini2026.01—9697.2
No MemoryModel=GPT-4o-mini2026.01—86—
Pass@10Model=GPT-4o-mini2026.01——92.8
RAGModel=GPT-4o-mini2026.01—91.491.6
Self-RAGModel=GPT-4o-mini2026.01—89.189.8