TS Concepts - advantages of concept with brackets

Question

Could anybody explain me why this two (slightly) different code snippets work similarily?
Is there any advantage of using one over the other?

#1 One way

template<typename T>
concept bool Swappable = requires (T a, T b) {
    {std::swap(a, b)} -> void;
};

#2 Other way

template<typename T>
concept bool Swappable() {
    return requires(T a, T b) {
        {std::swap(a, b)} -> void;
    };
};

What's the reason the second example has Swappable with brackets and returns "requirements"?

Possible main:

int main() {
    static_assert(Swappable<int>, "NOT SWAPPABLE!"); // Compiles when #1
    // static_assert(Swappable<int>(), "NOT SWAPPABLE!"); // Compiles when #2

    return 0;
}