List your company

Open-Web Reasoning on GAIA Open-web via SERPER (test)

46.8Pass@4Qwen3-8B
21.73628.24334.7541.257Jun 15, 2026

Evaluation results

MethodLinks
Qwen3-8BSampling Method=DivInit (DI)2026.0646.8
Gemma3-12BSampling Method=DivInit (DI)2026.0645.2
Qwen3-4BSampling Method=DivInit (DI)2026.0644.9
Qwen3-8BSampling Method=Standard parallel sampling (S)2026.0641.6
Qwen3-4BSampling Method=Standard parallel sampling (S)2026.0638.7
Gemma3-12BSampling Method=Standard parallel sampling (S)2026.0638
Gemma3-12BSampling Method=DivInit (DI)2026.0635
Gemma3-12BSampling Method=Standard parallel sampling (S)2026.0634
Gemma3-12BSampling Method=DivInit (DI)2026.0631.6
Qwen3-8BSampling Method=DivInit (DI)2026.0630.2
Qwen3-4BSampling Method=DivInit (DI)2026.0629
Qwen3-8BSampling Method=DivInit (DI)2026.0628.2
Gemma3-12BSampling Method=Standard parallel sampling (S)2026.0628.2
Qwen3-4BSampling Method=DivInit (DI)2026.0627.8
Qwen3-8BSampling Method=Standard parallel sampling (S)2026.0626
Qwen3-8BSampling Method=Standard parallel sampling (S)2026.0625.2
Qwen3-4BSampling Method=Standard parallel sampling (S)2026.0623.7
Qwen3-4BSampling Method=Standard parallel sampling (S)2026.0622.7