5 days Ago By Firstpost
OpenAI is expanding to accommodate new services as it seeks to become an indispensable tool. The tech giant has launched a new model, GPT 4.1, with two mini versions, GPT-4.
1 mini and GPT-4.1 nano. The new model is touted to improve coding, instruction, and following, and it can be used better in longer contexts.
The ChatGPT maker said that the new models will only be available on OpenAI’s application programming interface (API). OpenAI said that GPT 4.1, GPT-4.
1 mini, and GPT-4.1 nano are the company’s most advanced GPT-4o models across the board. Thanks to enhanced context comprehension, these models can now handle as many as one million tokens — the basic units of data an AI processes.
Additionally, their knowledge base has been updated to include information up to June 2024. Internal studies have shown that GPT-4.1 exhibits a 21 per cent improvement as compared to GPT-4o, while the new model is 27 per cent better at coding than GPT-4.
5. CEO Sam Altman said, “Benchmarks are strong, but we focused on real-world utility, and developers seem very happy.” According to OpenAI, this new family of models runs at a significantly reduced cost compared to GPT-4.
5. The company also announced that the GPT-4.5 preview currently accessible through the API will be discontinued in July, as the new models deliver equal or better performance.
On Monday, Altman, in a conversation with TED curator Chris Anderson, said that ChatGPT’s user base has doubled in just a few weeks. “Something like 10 per cent of the world uses our systems, now a lot," he said, according to a Forbes report. OpenAI has seen a rapid surge in its user base over the past few months, especially since late last year.
By December 2024, the platform reported approximately 300 million weekly active users on ChatGPT. That number likely doubled last month following the launch of the company’s newest image generation tool..
