List your company

Agent Performance on VitaBench In-Store

32.25Avg@4NoisyAgent
14.8319.352523.87528.3975May 26, 2026

Evaluation results

MethodLinks
NoisyAgentbackbone=Qwen3-32B2026.0532.2546
GSPObackbone=Qwen3-32B2026.0530.544
GRPObackbone=Qwen3-32B2026.0530.2543
DAPObackbone=Qwen3-32B2026.0529.542
Qwen3-32Bbackbone=Qwen3-32B2026.052638
NoisyAgentbackbone=Qwen3-8B2026.052437
DAPObackbone=Qwen3-8B2026.0523.2536
GSPObackbone=Qwen3-8B2026.052335
GRPObackbone=Qwen3-8B2026.0522.7535
Qwen3-8Bbackbone=Qwen3-8B2026.0515.524