@@CodeWithAhsan Good to know! Thank you for the clarification. Do you know if possible to access Google Calender with Summary API? I believe it need to have Google Workspace account but I am wondering if there is a walkalound.
Hey, there were some braking changes from the Chrome team (since this is still in experimental stages). However, I have updated the code with the latest changes .Try it out now :)
How is this in any way useful for the average user. There is not a single "normal" user out there who is going to go through all this and then open some "Dev Tool" program in order to use the Ai. That is in no way at all useful for the average/normal user! I open the Edge browser and I have a copilot side panel without having to do any of this google crap. I don't know where Google's head is in all of this. They average/normal Chrome user is not going to EVER do all the stuff you just showed in order to get Gemini up and running as a part of Chrome. It's ridiculous. And the average/normal Chrome user has absolutely zero understanding or knowledge of react or any other programing language. I am at a complete loss as far as Google's strategy or who they are targeting as the user base......because it sure is not the average or normal Chrome user.
Great question. As I mentioned. This is an early preview inside the Chrome Browser (as an API) :) which is NEVER meant for “an average” user. It is supposed to be tested by power users for now and when this is available for the general audience, you can imagine a much easier way for people to be able to use it. Secondly, gemini nano’s JavaScript API (which I used in my React code) is for developers and not for general users. So I can imagine your frustration with the video :) But it isn’t intended for you, and I should probably have done a better job with the title.
@@CodeWithAhsan I do understand where you are coming from. I was just a little/a lot frustrated. It just blows my mind that Google seems so far behind that it can't even include a useful Ai in perhaps their #1 used product, chrome.
I have followed all the steps correctly, including using the correct Chrome version. In 'chrome://flags/', I enabled 'Optimization Guide on Device' by setting 'BypassPerfRequirement' to 'Enabled', but I still can't find the option 'Optimization Guide on Device' in 'chrome://components/'. Please help me resolve this.
Hey, there are some breaking API changes. Please have a look at the updates document: goo.gle/chrome-ai-dev-preview-index Breaking changes in Chrome Canary 131.0.6776.0+ for the Prompt API In short, run (await ai.languageModel.capabilities()).available; in your chrome inspector's console tab and the model should start downloading. I will update the github repository and the readme.
@@leoxu1780 Hey, there are some breaking API changes. Please have a look at the updates document: goo.gle/chrome-ai-dev-preview-index Breaking changes in Chrome Canary 131.0.6776.0+ for the Prompt API In short, run (await ai.languageModel.capabilities()).available; in your chrome inspector's console tab and the model should start downloading. I will update the github repository and the readme.
Looking forward to screenshots of your prompts and gemini-nano's answers 😄😄
theres no optimisation guide on device model no matter what i do
Thank you for the tutorial! Do you know if it's possible to generate image using prompt api?
Hi, you're welcome.
And no, Gemini Nano is a text-only model and does not support multiple modalities like image, audio, etc.
@@CodeWithAhsan Good to know! Thank you for the clarification. Do you know if possible to access Google Calender with Summary API? I believe it need to have Google Workspace account but I am wondering if there is a walkalound.
hey i tried running the demo and i'm getting TypeError: Cannot read properties of undefined (reading 'create') in console
Hey, there were some braking changes from the Chrome team (since this is still in experimental stages). However, I have updated the code with the latest changes .Try it out now :)
How is this in any way useful for the average user. There is not a single "normal" user out there who is going to go through all this and then open some "Dev Tool" program in order to use the Ai. That is in no way at all useful for the average/normal user!
I open the Edge browser and I have a copilot side panel without having to do any of this google crap. I don't know where Google's head is in all of this. They average/normal Chrome user is not going to EVER do all the stuff you just showed in order to get Gemini up and running as a part of Chrome. It's ridiculous. And the average/normal Chrome user has absolutely zero understanding or knowledge of react or any other programing language. I am at a complete loss as far as Google's strategy or who they are targeting as the user base......because it sure is not the average or normal Chrome user.
Great question.
As I mentioned. This is an early preview inside the Chrome Browser (as an API) :) which is NEVER meant for “an average” user. It is supposed to be tested by power users for now and when this is available for the general audience, you can imagine a much easier way for people to be able to use it.
Secondly, gemini nano’s JavaScript API (which I used in my React code) is for developers and not for general users. So I can imagine your frustration with the video :) But it isn’t intended for you, and I should probably have done a better job with the title.
You can also imagine free tools(chrome extensions or websites) being developed that use Gemini Nano and you get to use it for free. Even offline
@@CodeWithAhsan I do understand where you are coming from. I was just a little/a lot frustrated. It just blows my mind that Google seems so far behind that it can't even include a useful Ai in perhaps their #1 used product, chrome.
@dbreardon Do you have stupid?
I have followed all the steps correctly, including using the correct Chrome version. In 'chrome://flags/', I enabled 'Optimization Guide on Device' by setting 'BypassPerfRequirement' to 'Enabled', but I still can't find the option 'Optimization Guide on Device' in 'chrome://components/'. Please help me resolve this.
I encountered exactly the same thing. I don't know what to do
Hey, there are some breaking API changes. Please have a look at the updates document:
goo.gle/chrome-ai-dev-preview-index
Breaking changes in Chrome Canary 131.0.6776.0+ for the Prompt API
In short, run (await ai.languageModel.capabilities()).available; in your chrome inspector's console tab and the model should start downloading.
I will update the github repository and the readme.
@@leoxu1780 Hey, there are some breaking API changes. Please have a look at the updates document:
goo.gle/chrome-ai-dev-preview-index
Breaking changes in Chrome Canary 131.0.6776.0+ for the Prompt API
In short, run (await ai.languageModel.capabilities()).available; in your chrome inspector's console tab and the model should start downloading.
I will update the github repository and the readme.