Do embedding vectors introduce randomness in output image?

Hi Everyone,

I understand that the model uses noise and that different noise will create different outputs. In Mid Journey I can set the seed and I get the same image with the same prompt.

If I set the seed and vary the prompt slightly – dark vs light background (see prompt below). I get different owl images. Why is this, is it due to the context embedding?

prompt A: fuzzy owl, light background --seed 123456**
prompt B: fuzzy owl, background background --seed 123456**

Many thanks!