To make the product widely accessible for developers, OpenAI said the new model comes with a more affordable pricing structure than its predecessor GPT-4.
In a groundbreaking move, OpenAI has announced the launch of a cutting-edge platform that allows users to create personalized versions of the popular AI chatbot ChatGPT, which made history months after its debut as the fastest-growing consumer app.
The company made the announcement during the DevDay conference hosted in San Francisco on November 6.
No Coding Experience Needed
According to the company’s blog post, the new platform caters to a diverse range of users, offering effortless customization without coding expertise.
“Anyone can easily build their own GPT – no coding is required. You can make them for yourself, just for your company’s internal use, or for everyone,” the company said.
Aside from the new feature that allows customized creation of personal versions of ChatGPT, the company said it has made a series of other improvements and new additions to its product offerings.
OpenAI also released a new assistant API that allows developers to build their own AI assistants that can offer a wide range of services with the ability to call models and tools.
OpenAI Unveils GPT-4 Turbo
During the event, OpenAI also unveiled the launch of its new AI model dubbed GPT-4 Turbo.
The GPT-4 Turbo, equipped with an extended context window and an updated knowledge base stretching to April 2023, promises to revolutionize the AI landscape.
Unlike the company’s older AI versions, with a knowledge base of September 2021 for GPT-3 and January 2022 for GPT-4, the latest model could provide answers to every event that occurred last year up until April this year.
OpenAI CEO Sam Altman said at the event on Monday that he was annoyed that the company’s previous creations had limited knowledge about the world.
“We are just as annoyed as all of you, probably more, that GPT’s knowledge about the world ended in 2021,” said Altman.
In addition to the expanded knowledge base, the GPT-4 Turbo brings forth many other enhancements, including expanded input capabilities of up to 300 pages and the integration of DALL-E 3 AI-generated images.
It is also developed with text-to-speech functionalities with six preset voices to choose from, allowing users to decide to hear the answer to a query in various voices.
OpenAI New Model Cheaper than GPT-4
To make the product widely accessible for developers, OpenAI said the new model comes with a more affordable pricing structure than its predecessor GPT-4.
The latest iteration aims to enhance the user experience while ensuring significant cost savings for developers, with input tokens now priced at $0.01 and output tokens at $0.03. The company explained:
“We also optimized its performance so we are able to offer GPT-4 Turbo at a 3x cheaper price for input tokens and a 2x cheaper price for output tokens compared to GPT-4.”
Paying developers can also test GPT-4 Turbo by using the gpt-4-1106-preview in the API. The company plans to introduce a stable production-ready model in the upcoming weeks.
Additionally, OpenAI said GPT-4 Turbo demonstrates improved performance on tasks that demand precise adherence to instructions, such as generating specific formats like responding in XML. It also supports a new JSON mode that ensures the model’s responses adhere to the valid JSON format.
The introduction of the response_format API parameter allows developers to restrict the model’s output, ensuring the generation of accurately structured JSON objects. This feature benefits developers utilizing JSON in the Chat Completions API for purposes beyond function calling.
Chimamanda is a crypto enthusiast and experienced writer focusing on the dynamic world of cryptocurrencies. She joined the industry in 2019 and has since developed an interest in the emerging economy. She combines her passion for blockchain technology with her love for travel and food, bringing a fresh and engaging perspective to her work.
