GUI Navigation on MMG2Skill-Bench GUI
77.25Success RateMMG2Skill
Evaluation Results
| Method | Links | |
|---|---|---|
| MMG2SkillBackbone=Kimi-K2.6, Prompting Variant=MMG2Skill2026.06 | 77.25 | |
| Raw GuideBackbone=Kimi-K2.6, Prompting Variant=Raw Guide2026.06 | 74.75 | |
| VanillaBackbone=Kimi-K2.6, Prompting Variant=Vanilla2026.06 | 69.75 | |
| MMG2SkillBackbone=Claude-Opus-4.6, Prompting Variant=MMG2Skill2026.06 | 64.75 | |
| VanillaBackbone=Claude-Opus-4.6, Prompting Variant=Vanilla2026.06 | 62.22 | |
| MMG2SkillBackbone=Claude-Sonnet-4.6, Prompting Variant=MMG2Skill2026.06 | 59.76 | |
| Raw GuideBackbone=Claude-Opus-4.6, Prompting Variant=Raw Guide2026.06 | 59.75 | |
| MMG2SkillBackbone=GPT-5.5, Prompting Variant=MMG2Skill2026.06 | 57.24 | |
| VanillaBackbone=Claude-Sonnet-4.6, Prompting Variant=Vanilla2026.06 | 49.76 | |
| Raw GuideBackbone=Claude-Sonnet-4.6, Prompting Variant=Raw Guide2026.06 | 47.5 | |
| MMG2SkillBackbone=Qwen3.6-Plus, Prompting Variant=MMG2Skill2026.06 | 39.99 | |
| VanillaBackbone=GPT-5.5, Prompting Variant=Vanilla2026.06 | 39.74 | |
| Raw GuideBackbone=GPT-5.5, Prompting Variant=Raw Guide2026.06 | 37.24 | |
| MMG2SkillBackbone=Gemini-3.1-Pro-Preview, Prompting Variant=MMG2Skill2026.06 | 35 | |
| Raw GuideBackbone=Gemini-3.1-Pro-Preview, Prompting Variant=Raw Guide2026.06 | 19.99 | |
| VanillaBackbone=Gemini-3.1-Pro-Preview, Prompting Variant=Vanilla2026.06 | 19.97 | |
| Raw GuideBackbone=Qwen3.6-Plus, Prompting Variant=Raw Guide2026.06 | 17.5 | |
| VanillaBackbone=Qwen3.6-Plus, Prompting Variant=Vanilla2026.06 | 15 |