Hello,
I use https://gemini.google.com user interface to upload a pdf file with a picture of a document and I get very good results there in reading and understanding the details of the document. When I use the same model in Google AI Studio I don’t get such good results anymore (there are problems with understanding more complex elements). Do you know how to get the same effect in Google AI studio / via API as in https://gemini.google.com user interface? I think it might be a matter of vision model configuration, but in Google AI studio I don’t see any options related to this
Thank you in advance for any help!
1 Like
In the Google AI Studio write the prompt. You try that. My suggestion 
1 Like
I’m using the same prompt in https://gemini.google.com and in Google Ai Studio and I’m getting different results… https://gemini.google.com have slighty better results. What to do to get the same result in Ai studio?
1 Like
Google AI Studio ana sayfasında sol yukarıda prompt yaz bölümü var buraya talep etmiş olduğun ve istediğin yanıtları alabilmek için prompt yazabilirsin bu sayede istediğin sonucu alabilirsin. Kesin çözüm değil ama bence deneyebilirsin
1 Like