Why Gemini API documentations and SDK is complicated as compared to OpenAI, Mainly in stream

I don’t know why but I found that Gemini Native SDK @google/genai is bit more complicated specially for newbies in this field as compared to OpenAI documentation, even Gemini native SDK has very complexity to provide files and images for understanding. Please Google fix this issue as I am loving your 2.0 models but my application requires stream options and your documentations is really making it very typical to to stream properly with all features including History, Files/Images, Instructions, Configs etc. But currently I am preffering OpenAI SDK to use Gemini Models but still it I am unable to find how to provide “Files/Docs (Like PDF, TXT, etc.)” with OpenAI SDK. Please help someone regarding this.