Openai Today stated that they will release a new family Artificial intelligence Models are optimized for superior encryption, as it enhances efforts to fight the increasingly fierce competition from companies like Google and Anthropic. Models are available for developers via Openai's application programming interface (API).
Openai is releasing three sizes of models: GPT 4.1, GPT 4.1 Mini and GPT 4.1 Nano. Kevin Weil, Product Manager at Openai, said on a livestream that the new models are better than the most widely used model of Openai, GPT-4O, and better than its largest and most powerful model, GPT-4.5, somehow.
GPT-4.1 has reached 55 percent on Swe Bench, a standardized benchmark used to measure the capacity of encryption models. The score is a higher percentage score than other Openai models. The new models are great people in encryption, they are great in complex instructions, they are great for construction agents, according to Weil Weil.
The ability for AI models and code editing has been significantly improved in recent months, allowing more automatic models and improving the ability to be called AI AGENT. Rivals like Humanity And Google Both introduce special models that are good at writing code.
The appearance of GPT-4.1 has been widely rumored for weeks. Openai clearly tested the model on a number of popular charts according to Alpha Quasar, the sources said. Some users of the stealth model ” report Impressive encryption ability. After that, Quasar fixed all the open issues that I encountered with another code (SIC) through the incomplete LLMS, a writer on Reddit.
All new models can analyze the code eight times at the same time, which helps improve the ability to improve and fix errors. The new models are also better in the instructions given by users, reducing the need to repeat commands in different ways to get the desired results. Openai has shown the demo of GPT-4.1 to build various applications including Flashcard applications to learn language.
The developers of developers are very much encrypted and we have improved the ability to write the functional code of the model, Mr. Michelle Michelle Pokrass, who worked after training at Openai, said in the second live broadcast. We worked to make it follow different formats and better discover repos, run units tests and record the translation code.
GPT-4.1 is 40 % faster than GPT.4O, Openai's most widely used model for developers. The cost of users' query has dropped by 80 % in this latest version, Openai said.
Today, Varun Mohan, CEO of Windsurf, a common tool for AI Codd, said the company tested GPT-4.1 and found that the new model was 60 % better than GPT-4O at its own benchmark. We find that GPT-4.1 has fewer cases of degeneration.
Over the past few years, Openai has caused a fever ChatgptA notable chatbot The first time was announced at the end of 2022In a developing business, it is more advanced to sell access to chatbot and AI models. In a TED interview last week, Altman said that Openai had 500 million users operating weekly and the use was grown very fast.