Applied AI·How models work
the number is billions of parameters, and the well-trained 8B beat the older 70B you assumed would win.
Model size naming (7B, 70B)
Draft summary, pending review
The number is the parameter count in billions. Bigger generally means more capable and slower, but the relationship is loose: a well-trained modern 8B can beat a poorly trained older 70B. For MoE models you will see 'total' vs 'active' parameters.