OpenAI, one of the most important companies in the field of artificial intelligence, which develops Chatgpt and generally, opened the GPT-4.1 artificial intelligence family to paid subscribers.
The new artificial intelligence model GPT-4.1, located on the GPT-4O, which was launched on May 13, 2024, is divided into three multimodal ”(text alongside text and visual perception) like all LLMs and all LLMs. In addition to the main version, the GPT-4.1, which also has a smaller-size “mini” and “nano” versions as it can be understood from the main version, brings the performance bar further up in almost every field, as there will be no surprise. OpenAI, who made the description of the GPT-4.1 Nano, the smallest, fastest and cheapest model so far, ısı, has introduced Mini with a lower parameter than the main model. At the first exit, OpenAI CEO Sam Altman said in a statement from the X account that the GPT-4.1 family would be presented only on the first stage of API. The company, which has been expected today, announced that it has opened its new model family to Plus, Pro and Team subscribers. Detailed performance data of the models can be found here.
You may be interested in
OpenAI CEO Sam Altman last month, “Capacity shortage may cause different problems.” warned. OpenAI, which is constantly increasing the number of users, cannot grow high demands. In his last statement on this issue, indicating that they have taken control of the situation for now, but new services and features may be delayed Sam Altman, due to capacity problems due to some problems may arise and services can start to work slower than usual, he said.
“We work as fast as we can to speed up things.” Altman says, server systems using the special GPUs from Nvidia. Technology giants such as Meta, X, Microsoft, Amazon and Google, GPU from Nvidia, can not enlarge the server systems at the desired level, OpenAI, in the face of intense user demand in recent years, “GPUs are melting/burning,” he says.