Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Visual Navigation on TIR-Bench Maze
Loading...
65
Accuracy
Qwen3-VL + V-ABS
8.528
23.189
37.85
52.511
May 11, 2026
Accuracy
Updated 21d ago
Evaluation Results
Method
Method
Links
Accuracy
Qwen3-VL + V-ABS
Backbone=Qwen3-VL-8B,...
2026.05
65
Intern-VL3-8B + V-ABS
Backbone=Intern-VL3-8B...
2026.05
56.7
Qwen2.5-VL + V-ABS
Backbone=Qwen2.5-VL-7B...
2026.05
46.7
Intern-VL3-8B
Backbone=Intern-VL3-8B
2026.05
26.7
GPT-4o + V-ABS
Backbone=GPT-4o, Strat...
2026.05
23.3
GPT-4o + VisuoThink
Backbone=GPT-4o, Strat...
2026.05
20.1
Qwen3-VL-8B
Backbone=Qwen3-VL-8B
2026.05
17.5
GPT-4o
Backbone=GPT-4o
2026.05
17.5
Qwen2.5-VL-7B
Backbone=Qwen2.5-VL-7B
2026.05
10.7
Feedback
Search any
task
Search any
task