Skip to content

Forum in maintenance, we will back soon 🙂

Notifications
Clear all

ChatGPT

6 Posts
4 Users
1 Reactions
82 Views
(@google-rayazsiddiqi)
Posts: 95
Estimable Member
Topic starter
 

All the examples shown in the course from prompt engineering are gpt 3.5 turbo, why is this? If I want to use  gpt4 what should the code be and what benefit would 4 give me? Thanks

 
Posted : 05/21/2024 4:18 pm
SSAdvisor
(@ssadvisor)
Posts: 1139
Noble Member
 

@google-rayazsiddiqi hello Ray, would need to change the OpenAI model in the code. There are pros and cons for each model that you need to weigh based on your application.

Regards,
Earnie Boyd, CEO
Seasoned Solutions Advisor LLC
Schedule 1-on-1 help
Join me on Slack

 
Posted : 05/21/2024 4:48 pm
(@google-rayazsiddiqi)
Posts: 95
Estimable Member
Topic starter
 

@ssadvisor Thanks How do I decide which one to use?

 
Posted : 05/21/2024 8:43 pm
SSAdvisor
(@ssadvisor)
Posts: 1139
Noble Member
 

@google-rayazsiddiqi the amount of data (tokens) you need to meet the criteria for your project.

Regards,
Earnie Boyd, CEO
Seasoned Solutions Advisor LLC
Schedule 1-on-1 help
Join me on Slack

 
Posted : 05/22/2024 12:02 am
Hasan Aboul Hasan
(@admin)
Posts: 1208
Member Admin
 

A good practice is to develop your app, tool, workflow, or whatever you are creating and use the most powerful model first, like gpt-4 or 4o, which was released last week.

Then, when you get the results you want, you can try using a cheaper model and see if you get the same results. 

 

 
Posted : 05/22/2024 9:27 am
(@husein)
Posts: 464
Member Moderator
 

@google-rayazsiddiqi Each Model has its own pros and cons, For example, GPT-4 usually gives better results since its trained on more data but it costs more to use. Unlinke, GT-3.5 which also gives good results but not like GPT-4 and its way cheaper.

 
Posted : 05/22/2024 9:50 am
Share: