sagarpatil

o1 models might use multiple methods to come up with an idea, only one of them might be correct, that's what they show in ChatGPT. So it just summarises the CoT, does not include the whole reasoning behind it.

With Google's 1 million token and Sonnet 3.5's 200,000 token limit, is there any advantage of using this over just uploading the pdf files and ask questions about it. I was under the impression that you will get more accurate results by adding the data in chat.

IANAL?

Off topic: How does the judiciary decide which judge to choose for such highly technical case?

You are 100% right about using unified diffs to overcome lazy coding. Cursor.sh has also implemented unified diffs for code generation. You ask it to refactor code, it writes your usual explanation but there's a apply diff button which modifies the code using diff and I've never seen placeholder code in it.

A chat plugin for VS Code will be a game changer.

There are YouTube videos that go into detail. From what I can remember, it first creates an embedding of your full code, it then refers to your open file and the files next to your current tab, it then extracts the most useful code related to your question.

I asked the founder this question previously and if I remember it correctly, they said they don't have any plans for an API.

It's not you. It's high interest rate. In a bid to get the inflation down, the fed has been increasing the interest rates. Currently, they are at a 22 years high. Unemployment rate is at 3.7%. Most of the big tech companies have been doing a mass layoff last and this year. The job market has to absorb them too.

GPT-4-1106-preview aka turbo has 128k tokens. Are they saying GPT-4 (0613) with 32k is better than GPT-4 Turbo?

There's no need to be rude.

Congratulations on the launch. Looks great! A couple of questions if you don't mind me asking:

1. Why did you build this tool when we have thousands of time zone comparison websites, apps and inbuilt functionality in smartphone time app?

2. Google already does this natively

3. It will be really tough to rank this on Google

4. Do you plan on monetizing this?

Bye bye Quora!

Have you tried perplexity.ai? It's like ChatGPT and Google had a baby. Looks very promising and I'm seeing a lot of tech leaders (example Toby of Shopify) moving to it.

Till the time they fix it, there's this add-on which fixes the issue: https://addons.mozilla.org/en-US/android/addon/google-search...

I switched to firefox nighly in Android and I don't think I'm going back to Chrome. The amount of ads, tracking and the overall experience was getting out of hand.

Yep. That's the most interesting thing here. OP can you elaborate on how you got it to develop a full fledged application?

Asked it to write code for two python programs (one of them tricky) and it got both of them right in the first pass. Looks promising.

This is awesome. Congratulations. I wanted to introduce my kids to programming so this is very timely. Thanks for building it.

What a sh*t show. For someone who is building on top of OpenAI this is very unnerving. My twitter is full of heart emojis and OpenAI is nothing without its employees.

How does one go about hosting this and using this as an API?

I'm impressed. What did your dad do to help you get interested in coding?

Yes. It's making around 150/month through affiliate ads. My API expenses are around $50. I've listed the plugin for sale on Acquire.com as I don't have time to work on it.

You could cook something like this using Microsoft Autogen. It does allow you to daisy chain the models.

I've been using GPT-4 for coding since it's launched. I did try Phind when you launched it with GPT4 and found it useful but I use VS Code and hate having to switch. I see you do have a VS Code extension now (great job!). I tried the phind model and at first glance it definitely looks better than GPT-4 wrt coding. Do you have any plans to provide this model as an API?

Unfortunately, there's no direct link. You need a ChatGPT Plus subscription. Go to plugin store and search for Aardvark AI.

You are right about the context window limitation. I exclusively use Azure OpenAI GPT-4 32k version and it's been a game changer when coding on complex projects.

This is brilliant. I also created a ChatGPT plugin using only GPT-4. Aardvark AI (Search for Products on Google Shopping in Real-Time, No Ads, Only Trusted Stores) It current gets around 6000-8000 search queries each month.

BTS: It uses Google Shopping API to get real time results.

Now returns a 404. Interesting.

Azure API is definitely faster than OpenAI and they also seem to provide access to 32k generously compared to OAI.