Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

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

46.8Pass@4

Qwen3-8B

21.73628.24334.7541.257Jun 15, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.06
46.8
2026.06
45.2
2026.06
44.9
2026.06
41.6
2026.06
38.7
2026.06
38
2026.06
35
2026.06
34
2026.06
31.6
2026.06
30.2
2026.06
29
2026.06
28.2
2026.06
28.2
2026.06
27.8
2026.06
26
2026.06
25.2
2026.06
23.7
2026.06
22.7