#here is what I built.
1 messages · Page 1 of 1 (latest)
Cool idea :).
Unfortunately, I can't get it to work for me. In the LLM options I only see “OpenAI”, “ollama” or “Google”. And I need an API key for Google. However, this should not be necessary for the PromptAPI.
Is it possible that you are not using the on-device APIs, but the Gemini API?
yes, that is correct you need a key. its not the in-chrome prompt, it is the google prompt api
you can get an api key from here
https://aistudio.google.com/apikey
I checked very early on to make sure this implementation was within the acceptable bounds for this competittion
(just saw your last sentence, yes, Gemini API)
Hmm🤔. Maybe @trim elk or @real sable can clarify this. But I am pretty sure you have to use the built-in PromptAPI (the one that runs locally on the users device, not in the cloud) to qualify for the challenge..
The contest rules say that its not limited to the built-in api. I asked this about a month ago and got a confirmation. But I am standing by
Yes thats true. In addition to the built-in APIs (summarizer, PromptAPI, Writer, Rewriter, etc), you can use other APIs. But I am pretty sure you have to use at least one of them..
here is what I Built, assitant after scanning a barcode on product https://github.com/kurodohenroonsen/Barcode-Translator
@twin mural I'm trying your ask my website but it is realllllyyy slooooowwwwwwww on my macbook pro 2,3 GHz Quad-Core Intel Core i7 Intel Iris Plus Graphics 1536 MB