Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
GUI Navigation on MMG2Skill-Bench GUI
Loading...
77.25
Success Rate
MMG2Skill
12.51
29.3175
46.125
62.9325
Jun 1, 2026
Success Rate
Updated 1d ago
Evaluation Results
Method
Method
Links
Success Rate
MMG2Skill
Backbone=Kimi-K2.6, Pr...
2026.06
77.25
Raw Guide
Backbone=Kimi-K2.6, Pr...
2026.06
74.75
Vanilla
Backbone=Kimi-K2.6, Pr...
2026.06
69.75
MMG2Skill
Backbone=Claude-Opus-4...
2026.06
64.75
Vanilla
Backbone=Claude-Opus-4...
2026.06
62.22
MMG2Skill
Backbone=Claude-Sonnet...
2026.06
59.76
Raw Guide
Backbone=Claude-Opus-4...
2026.06
59.75
MMG2Skill
Backbone=GPT-5.5, Prom...
2026.06
57.24
Vanilla
Backbone=Claude-Sonnet...
2026.06
49.76
Raw Guide
Backbone=Claude-Sonnet...
2026.06
47.5
MMG2Skill
Backbone=Qwen3.6-Plus,...
2026.06
39.99
Vanilla
Backbone=GPT-5.5, Prom...
2026.06
39.74
Raw Guide
Backbone=GPT-5.5, Prom...
2026.06
37.24
MMG2Skill
Backbone=Gemini-3.1-Pr...
2026.06
35
Raw Guide
Backbone=Gemini-3.1-Pr...
2026.06
19.99
Vanilla
Backbone=Gemini-3.1-Pr...
2026.06
19.97
Raw Guide
Backbone=Qwen3.6-Plus,...
2026.06
17.5
Vanilla
Backbone=Qwen3.6-Plus,...
2026.06
15
Feedback
Search any
task
Search any
task