List your company

Game Interaction on MMG2Skill-Bench Game

83.33Success RateMMG2Skill
7.066826.865946.66566.4641Jun 1, 2026

Evaluation results

MethodLinks
MMG2SkillBackbone=Claude-Opus-4.6, Prompting Variant=MMG2Skill2026.0683.33
MMG2SkillBackbone=GPT-5.5, Prompting Variant=MMG2Skill2026.0680
Raw GuideBackbone=GPT-5.5, Prompting Variant=Raw Guide2026.0676.67
VanillaBackbone=GPT-5.5, Prompting Variant=Vanilla2026.0673.33
MMG2SkillBackbone=Claude-Sonnet-4.6, Prompting Variant=MMG2Skill2026.0673.33
VanillaBackbone=Claude-Opus-4.6, Prompting Variant=Vanilla2026.0666.67
Raw GuideBackbone=Claude-Opus-4.6, Prompting Variant=Raw Guide2026.0666.67
MMG2SkillBackbone=Gemini-3.1-Pro-Preview, Prompting Variant=MMG2Skill2026.0666.67
Raw GuideBackbone=Claude-Sonnet-4.6, Prompting Variant=Raw Guide2026.0660
VanillaBackbone=Claude-Sonnet-4.6, Prompting Variant=Vanilla2026.0656.67
MMG2SkillBackbone=Kimi-K2.6, Prompting Variant=MMG2Skill2026.0646.67
MMG2SkillBackbone=Qwen3.6-Plus, Prompting Variant=MMG2Skill2026.0646.67
VanillaBackbone=Gemini-3.1-Pro-Preview, Prompting Variant=Vanilla2026.0633.33
Raw GuideBackbone=Gemini-3.1-Pro-Preview, Prompting Variant=Raw Guide2026.0623.33
VanillaBackbone=Kimi-K2.6, Prompting Variant=Vanilla2026.0616.67
Raw GuideBackbone=Kimi-K2.6, Prompting Variant=Raw Guide2026.0616.67
VanillaBackbone=Qwen3.6-Plus, Prompting Variant=Vanilla2026.0616.67
Raw GuideBackbone=Qwen3.6-Plus, Prompting Variant=Raw Guide2026.0610