It's the screen recording software, which shows during the edit. It doesn't actually change the cursor :/

Don't think it works on Ventura, would have to update to Sonoma.

Try downloading through the link.

They are slowly rolling out to all users. OpenAI is rolling out to paid subscribers first. If you have the free version, it might take some weeks.

@@SchoolofMachineLearning I have the paid version and still says no access yet... It's probably also something to do with regions

Sell it buy new one

@@coinbot8248 are u going to give me that money , MF

@@coinbot8248 I feel ashamed to reply you because it is hard to explain someone who has only 2 brain cells .

Yes it is the official app

The voice feature is currently not available, OpenAI has said they will release it in the next couple of weeks.

@@SchoolofMachineLearningyeah i know! But in my app for Mac not show icon microphone 😮. And the version my app from Mac is 1.2024.134 is the last version?

In an upcoming video I will show how to use the API :)

@SchoolofMachineLearning thank you, I have already seen lots of videos on the subject but as I have no level in coding so I can't understand

@SchoolofMachineLearning could you explain to me how to use APIs I had ideas but I don't know if they are feasible. and above all I watch tutorials but the KZbinrs use terms that I don't understand and they don't explain in detail so since I don't know anything about coding I have 1000 questions each time I am interested in the creation of GPT and at the same time SEO, subjects that interest me there are several and questions on each of these subjects I have dozens, when I watch a tutorial to learn or develop my skills I always want on average to interrupt the KZbinr to ask him questions, so this is what I would like to do thanks to artificial intelligence and I am disappointed given that gpt4 was sold to me as being much more intelligent and yet the results it gives me when I give it the transcript of a video and that I ask him to redo the tutorial for me by addressing someone at my level the results are as hollow as with the free version but I understood that by creating GPT I could thus make the chat a lot more intelligent? but I have a lot of questions to understand how to do my gpt correctly and above all I think that if I understood how to use the APIs and have use cases it could help me answer a lot of my questions by myself

Are you using an Intel MacBook?

I think the current version of the Mac app might not be available for intel MacBooks. OpenAI didn't mention anything related to that AFAIK, might release an intel version later on.

My pleasure!

What error do you get?

@@SchoolofMachineLearning l have sonomo 14.5 its says not compatible with mac

There is ChatGPT app for Android on the Play Store: play.google.com/store/apps/details?id=com.openai.chatgpt&hl=en_CA&gl=US&pli=1

Windows has Copilot, which you can use now but yeah have to wait.

me wanna talk to humannn robotttt

Copilot does run on GPT4, think they might have upgraded to GPT4o as well. But yeah I get your point.

@@SchoolofMachineLearning Yes its running on 4 or 4o. Paid users getting more fetches than free user of course. But it's just too slow in my opinion. I would also like more control over where it goes on my screen, atleast which monitor its on.

@@SchoolofMachineLearning I see your Ollama videos.. tbh I prefer to run ollama locally than copilot online.

I'm running M1 Pro on Sonoma (14.4.1) and it's working fine.

Afaik it has been rolled out to all +subscribers, you should get it soon I guess.

not possible to download on windows yet, they said they will add it later on in the year.

Not currently available for Windows laptop. For Android you can download via Play Store: play.google.com/store/apps/details?id=com.openai.chatgpt&hl=en_CA&gl=US

Currently they don't have an option for 3 minute songs.

I'm not sure what you mean. After downloading ChatGPT-4o, and opening it, what message did you get?

Is not official app from OpenAI

only for premium users now, eventually will be free chatgpt4o

it's the official app from OpenAI. if there are issues with login, it might just be that OpenAI did not enable it for your account yet. Think in the next couple of weeks, all free users should be able to directly download it as well.

haha :)

Glad you liked it

thank you :)

Yes, you can. Here is a tutorial for a nice interface using webUI: github.com/open-webui/open-webui. You can also directly use on Meta.ai.

Thanks, fixed!

thank you :)

Meta doesn't directly provide an API access but you can access via Groq/Replicate/Microsoft/Databricks.

open webui already exists too

For what you require it makes more sense to call Llama via an API as it will be much cheaper. It's currently $0.64/1M input and $0.80/1M output tokens on Groq AI (that's the cheapest one I've seen). For hardware, I haven't built anything like that so not sure, maybe an A100? :D But for a single user from what I've seen online, good specs are: An Apple M2 Ultra w/ 24-core CPU, 60-core GPU, 128GB RAM (costs $8000 with the monitor) runs Meta-Llama-3-70B-Instruct.Q4_0.llamafile at 14 tok/sec (prompt eval is 82 tok/sec).

@@SchoolofMachineLearning the sort of application I am considering requires an onpremise deployment so deploying it in cloud/consuming via api isn't an option. I am a bit more inclined towards Linux/Windows ecosystem. What would be the total VRAM/Ram required for the 70b model. Also does using 4bit quantized model result in some loss of accuracy, is that noticeable in the output?

Meta doesn't provide llama 3 API directly afaik but if you want to try out llama 3 you can do so on Meta.ai. A lot of other companies provide llama 3 API such as Databricks, Replicate, Microsoft, etc.

groq offre llma 3 70 b for free with api

It will have extremely poor performance, even then I don't think you will be able to run. But you can give it a shot.

I tried it and it's a waste of time. Computer freezes with simple commands and takes ages to come out of freeze. m3 macbook pro with 8gb ram

The short answer is to get a more powerful GPU :D

​@@SchoolofMachineLearningwhat should be the minimum GPU? Is RTX 3060 12GB enough?

I don't think that is going to be enough. By default, Ollama downloads a 4-bit quant. Which for Llama 3 70B is 40 GB. Your GPU has only 12 GB of VRAM, so the rest has to be offloaded into system RAM, which is much slower. You have two options: - Use the 8B model instead (ollama run llama3:8b) - Use a smaller quant (ollama run llama3:70b-instruct-q2_K)

I ran into the same, and having looked around it appears £20-30K GPUs with ~40GB VRAM are the type you'd need to manage the 70b model. It is, after all, 40GB of data; where your GPU is insufficient, this will be loaded to your RAM, which is exponentially slower than video card memory at performing this work.

@@SchoolofMachineLearning what is that q2_K? i have a 12GB 3080Ti, is that the best option for me? I read some who attempted this found the 7b model was superior.

The requirements are: - 16GB memory for 8B model. - 32GB memory for 70B model (even then it is very slow). I have not tried the 70B model on my laptop but I'm assuming it is almost not usable.

@@SchoolofMachineLearning How can I check memory requirements information about Llama-3 models? Especially I want to know the requirements for 70B model.

70B model on 32GB mac m1 max is taking like a minute per word... 8B model is very fast.