List your company

Multi-turn Agent Task on Search-QA

50.49Success RateGRPO
13.060422.777732.49542.2123Jun 26, 2026Jun 27, 2026Jun 28, 2026Jun 29, 2026Jun 30, 2026

Evaluation results

MethodLinks
GRPOStudent Model=None, Teacher Model=Qwen3-30B-A3B GRPO2026.0650.492.65
ATODStudent Model=Qwen3-4B, Teacher Model=Qwen3-30B-A3B GRPO2026.0649.122.62
OPDStudent Model=Qwen3-4B, Teacher Model=Qwen3-30B-A3B GRPO2026.0648.932.69
SODStudent Model=Qwen3-4B, Teacher Model=Qwen3-30B-A3B GRPO2026.0648.582.61
TCODStudent Model=Qwen3-4B, Teacher Model=Qwen3-30B-A3B GRPO2026.0648.392.57
GRPOStudent Model=None, Teacher Model=Qwen3-4B GRPO2026.0648.192.51
GRPOStudent Model=Qwen3-4B, Teacher Model=None2026.0648.192.51
TRIAGEModel=Qwen2.5-7B-Instruct, Judge=Qwen3-8B-thinking, Evidence=true2026.0648.1—
SDARStudent Model=Qwen3-4B, Teacher Model=None2026.0647.82.59
TRIAGEModel=Qwen2.5-7B-Instruct, Judge=Qwen3-8B-thinking, Evidence=false2026.0646.4—
ATODStudent Model=Qwen3-1.7B, Teacher Model=Qwen3-4B GRPO2026.0645.212.64
TCODStudent Model=Qwen3-1.7B, Teacher Model=Qwen3-4B GRPO2026.0645.022.74
TRIAGEModel=Qwen2.5-7B-Instruct, Judge=Qwen3-8B no-think2026.0645—
OPDStudent Model=Qwen3-1.7B, Teacher Model=Qwen3-4B GRPO2026.0644.732.66
SODStudent Model=Qwen3-1.7B, Teacher Model=Qwen3-4B GRPO2026.0644.432.69
GRPOModel=Qwen2.5-7B-Instruct2026.0643.3—
ATODStudent Model=Qwen3-0.6B, Teacher Model=Qwen3-4B GRPO2026.0642.332.77
TRIAGEModel=Qwen3-1.7B-Instruct, Judge=Qwen3-8B-thinking, Evidence=true2026.0642.3—
SDARStudent Model=Qwen3-1.7B, Teacher Model=None2026.0642.242.67
TCODStudent Model=Qwen3-0.6B, Teacher Model=Qwen3-4B GRPO2026.0642.042.67
GRPOStudent Model=Qwen3-1.7B, Teacher Model=None2026.0641.992.64
OPDStudent Model=Qwen3-0.6B, Teacher Model=Qwen3-4B GRPO2026.0641.462.77
TRIAGEModel=Qwen3-1.7B-Instruct, Judge=Qwen3-8B-thinking, Evidence=false2026.0641.1—
TRIAGEModel=Qwen3-1.7B-Instruct, Judge=Qwen3-8B no-think2026.0640.2—
SODStudent Model=Qwen3-0.6B, Teacher Model=Qwen3-4B GRPO2026.0640.192.81
SDARStudent Model=Qwen3-0.6B, Teacher Model=None2026.0639.62.43
GRPOModel=Qwen3-1.7B-Instruct2026.0639.4—
GRPOStudent Model=Qwen3-0.6B, Teacher Model=None2026.0639.362.69
VanillaStudent Model=Qwen3-4B, Teacher Model=None2026.0632.711.77
VanillaStudent Model=Qwen3-1.7B, Teacher Model=None2026.0631.152.09
VanillaStudent Model=Qwen3-0.6B, Teacher Model=None2026.0614.51.57