Per-model coverage (of 10 words)
Solved Failed play Missing
Kimi-K3
9/10
GLM-5.2
9/10
MiniMax-M3
7/8
MiMo-V2.5
4/10
Mistral-Med
4/10
Gemma-4-31B
7/10
gpt-oss-120b
7/10