OpenAI Introduces ChatGPT-4 Turbo With Cheaper Increases And Prices

OpenAI introduces ChatGPT-4 Turbo (photo: Unsplash)

JAKARTA - In March, OpenAI officially released the first version of the ChatGPT-4. Now, OpenAI is launching a preview of the next generation model, GPT-4 Turbo with event knowledge until April 2023.

In the first version, ChatGPT-4 is only able to provide answers related to events from data until September 2021. With the presence of the GPT-4 Turbo, it means that you can get more answers from the latest events.

The company claims that the GPT-4 Turbo has a 128k context window, so it can load the equivalent of more than 300 text pages in one command.

We're rolling out new features and improvements that developers have been asking for:1. Our new model GPT-4 Turbo supports 128K context and has fresher knowledge than GPT-4. Its input and output tokens are respectively 3× and 2× less expensive than GPT-4. It’s available now to…

— OpenAI (@OpenAI) November 6, 2023

We're rolling out new features and improvements that developers have been asking for:1. Our new model GPT-4 Turbo supports 128K context and has fresh knowledge than GPT-4. It's input and output tokens are respectively 3times and 2 times less exponential than GPT-4. It's available now to...

"We are also optimizing its performance so that we can offer the GPT-4 Turbo with 3x cheaper input token prices and 2x cheaper output token prices compared to the GPT-4," the company wrote on its official website, quoted Tuesday, November 7.

The input fee is only 0.01 US dollars (IDR 155 silver) per 1,000 tokens. Cheaper than 0.03 US dollars (IDR 467 silver) per token on the GPT-4.

GPT-4 Turbo will be available to try by all paying developers by continuing the gpt-4-1106-viewAPI. OpenAI also plans to release a stable production ready model in the coming weeks.

This latest version of GPT-4 will continue to receive image commands, text-to-speech requests, and integrate DALL-E 3, a feature that was first announced in October.

However, the upgrade to the GPT-4 Turbo means users can ask the model to perform more complex tasks at one time.

People can even tell the Turbo GPT-4 to specifically use the coding language of their choice to get results, such as the code in XML or JSON.

"We released several upgrades today, including the ability to call multiple functions in one message: users can send one message asking for some action, such as open car window and turn off AC', which previously required some action," OpenAI added.