#Hi want to know Gemini changed the model
1 messages · Page 1 of 1 (latest)
what do you mean? when you say that gemini changed the model?
im currently looking at the documentation and found this:
https://cloud.google.com/vertex-ai/generative-ai/docs/samples/generativeaionvertexai-gemini-all-modalities
Like gemini-1.5 flash model is working for images?? In the start Gemini give another model name: pro vision for image
So which model we can use for image???
That model was first deprecated, and then retired from service. Trying to use it gives a 404 now. The Gemini 1.5 (flash and pro) are the multimodal models that include vision
For Gemini flash I receive error for image description
Which error are you getting? It works for me.
i used this model="gemini-1.5-flash",
Eror
I am sorry, I cannot process or analyze images. I am a text-based chat assistant and do not have the capability to interpret medical reports from images.
Is that in AI Studio or using your own code?
The model doesn’t know its own capabilities, so you can’t trust the message it responds with.
it's my own
is there any problem in my code?
Right, to send an image, you can’t just put the image where previously you sent text. There is a type Part that can be text, or inlineData, or fileData. Start with the documentation here - https://ai.google.dev/gemini-api/docs/vision?lang=python
You can also use AI Studio first and then ask it to generate code. Then it will show you how to do it.