Optimizing turning integer vectors into binary vectors

I have a bunch of 4 million integer vectors that I have to convert to binary. My code is as follows:

def integer_vectors_to_binary(data, bits=16):
    bin_arr = []
    for arr in tqdm(data, desc="Processing", ncols=100):
        binary_array = [list(map(int, format(x, f'0{bits}b'))) for x in arr]
        bin_arr.append(np.array(binary_array).flatten())
    return np.asarray(bin_arr)

Now, the issue is that the process takes too long for my very large input set. Therefore, I was wondering if this code could be optimized. Here's a sample output:

vec_a = np.asarray([12, 15, 14])
print(integer_vectors_to_binary([vec_a]))

The output is:

array([[0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 0]])

Think of vec_a with 254 elements where each element is an integer with values from 0-65535.