List your company

Tool-use task completion on SkillCraft Hard

20Success RateMinimax-M2.1
6.489.9913.517.01Feb 28, 2026

Evaluation results

MethodLinks
Minimax-M2.1Mode=Skill2026.0220
Claude 4.5 SonnetMode=Base2026.0220
Claude 4.5 SonnetMode=Skill2026.0220
Minimax-M2.1Mode=Base2026.0218
GPT-5.2Mode=Skill2026.0217
Gemini 3 ProMode=Skill2026.0217
GPT-5.2Mode=Base2026.0216
Gemini 3 ProMode=Base2026.0216
DeepSeek-V3.2-EXPMode=Skill2026.0215
DeepSeek-R1Mode=Skill2026.0215
GLM-4.7Mode=Skill2026.0215
GLM-4.7Mode=Base2026.0212
DeepSeek-R1Mode=Base2026.0211
DeepSeek-V3.2-EXPMode=Base2026.029
Kimi-K2-ThinkingMode=Base2026.028
Kimi-K2-ThinkingMode=Skill2026.027