#Hey all, quick question! It seems like
1 messages ยท Page 1 of 1 (latest)
The tech industry has an unfortunate habit of poor naming conventions.
Don't confuse the following as they are all different:
- Gemini Applications: various user interfaces
- Gemini API: programming interface
- Gemini Models: such as Gemini 1.0 and Gemini 1.5
For image generation you would have to look at Imagen text-to-image diffusion model
See https://code.iaflw.com/2024/02/gemini-versus-gemini-understanding.html to untangle the terminology.
What makes the confusion worse is that google claimed that as a truly multimodal model, Gemini the model was able to generate images (back during the demo stuff)
Sure Google didn't do anyone any favors but this is not limited to Google.
ChatGPT != OpenAI API != GPT
Which is only slightly better as they didn't call it GPT API ๐คช
Google did the most painful one though
Gemini, the site at gemini.google.com
Gemini, the model
Gemini, renamed duet ai
Gemini, the assistant
At least openai didnt name everything chatgpt
Can't argue with that ๐คฃ
Then they say Gemini API when referring to the generativelanguage api
and then we have the Vertex AI API to access models including Gemini
๐๐
Hey this really helped me to understand! Thanks a lot
Thanks! it helped. It sad that I can't build anything generate image with API for Hackathon. There should be tons of other ideas for text based output. Thanks
Image generation is done using the Imagen model.
I'm not involved with nor participating in the hackaton in anyway but I see others requesting and getting access to Imagen through Vertex API. I don't know the conditions, restrictions, or potential costs.
Check out the hackathon forum at https://googleai.devpost.com/forum_topics
Amazing! Thanks for info
Good luck and have fun!
can we integrate - lets say google location api, search api and other apis on top of gemini for the hackathon ? even 3rd party api ?