Share your thoughts, 1 month free Claude Pro on usSee more

Tool-use task completion on SkillCraft Hard

20Success Rate

Minimax-M2.1

Updated 1mo ago

Evaluation Results

Method	Links
Minimax-M2.1 2026.02		20
Claude 4.5 Sonnet 2026.02		20
Claude 4.5 Sonnet 2026.02		20
Minimax-M2.1 2026.02		18
GPT-5.2 2026.02		17
Gemini 3 Pro 2026.02		17
GPT-5.2 2026.02		16
Gemini 3 Pro 2026.02		16
DeepSeek-V3.2-EXP 2026.02		15
DeepSeek-R1 2026.02		15
GLM-4.7 2026.02		15
GLM-4.7 2026.02		12
DeepSeek-R1 2026.02		11
DeepSeek-V3.2-EXP 2026.02		9
Kimi-K2-Thinking 2026.02		8
Kimi-K2-Thinking 2026.02		7