List your company

Multi-Agent System Performance on Database

71.72Task Success Rate (TS)MARBLE
66.811268.085669.3670.6344Mar 27, 2026

Evaluation results

MethodLinks
MARBLEModel=Qwen3-32B-Instruct2026.0371.7269.1569.68
A-MemModel=Qwen3-next-80B2026.0371.5970.3971.38
A-MemModel=Qwen3-32B-Instruct2026.0371.4373.2371.2
W.o. MemoryModel=Deepseek-v3.22026.037182.475.55
LLMA-MemModel=Deepseek-v3.22026.037181.0776.2
W.o. MemoryModel=Qwen3-32B-Instruct2026.037166.0767.87
LLMA-MemModel=Qwen3-next-80B2026.0370.4168.5569.98
LLMA-MemModel=Qwen3-32B-Instruct2026.0370.4176.1871.06
LLMA-MemModel=Claude-Sonnet-4.52026.037088.2778.52
MARBLEModel=Deepseek-v3.22026.037075.672.62
MARBLEModel=Claude-Sonnet-4.52026.0369.782.9675.7
A-MemModel=Claude-Sonnet-4.52026.0369.782.3175.13
W.o. MemoryModel=Claude-Sonnet-4.52026.0369.3988.0377.35
W.o. MemoryModel=Qwen3-next-80B2026.036972.3267.61
A-MemModel=Deepseek-v3.22026.0368.6973.4870.51
MARBLEModel=Qwen3-next-80B2026.036768.5364.38