Web Agent Navigation on TimeWarp Wiki Environment v1:6 benchmark leaderboard · SOTA2 Research

Web Agent Navigation on TimeWarp Wiki Environment v1:6

100Success RateHuman

Evaluation Results

Method	Links
Human2026.03		100
Qwen-3 4B (TimeWarp-BC)2026.03		31
Qwen-3 4B Thinking2026.03		27.2
Qwen-3 4B Thinking2026.03		26.9
Qwen-3VL 8B2026.03		20.8
Qwen-3 4B2026.03		20.3
Qwen-3VL 8B2026.03		19
Qwen-3 4B2026.03		19
Qwen-3 4B Thinking (TimeWarp-BC)2026.03		16.1
Gemma-3 12B2026.03		15.4
Gemma-3 12B2026.03		14
Llama-3.1 8B (TimeWarp-BC)2026.03		12.7
Qwen-3 4B2026.03		11.5
Qwen-3VL 8B Thinking2026.03		11.1
Qwen-3 4B2026.03		10.2
Gemma-3 12B2026.03		9.9
Qwen-3VL 8B Thinking2026.03		8.8
Gemma-3 12B2026.03		7.1
Qwen-3VL 8B2026.03		6.8
Qwen-3 4B Thinking2026.03		4.3
Qwen-3 4B Thinking2026.03		4.1
Llama-3.1 8B2026.03		2.3
Qwen-3VL 8B Thinking2026.03		1.4
Qwen-3VL 8B2026.03		0.7
Llama-3.1 8B2026.03		0.5
Qwen-3VL 8B Thinking2026.03		0
Llama-3.1 8B2026.03		0
Llama-3.1 8B2026.03		0