OpenAI introduced the GPT-4.1 family of AI models on Monday, which includes GPT 4.1, GPT 4.1 Mini, and GPT 4.1 Nano, reports The Indian Express.
These models will be available to developers in application programming interface (API) format.
The company claims that the GPT-4.1 models “outperform GPT 4o and GPT 4o Mini across the board,” with major improvements in coding capabilities and a longer context window of up to 1 million tokens, with a knowledge cutoff of June 2024.
OpenAI CEO Sam Altman in an X post on Monday, wrote,“GPT-4.1 (and Mini and Nano) are now available in the API! These models are great at coding, instruction following, and long context (1 million tokens). Benchmarks are strong, but we focused on real-world utility, and developers seem very happy. The GPT-4.1 family is API-only,”
“We’re bringing the latest GPT-4.1 models from OpenAI to Foundry, representing significant advancements in coding, instruction following, and long-context processing. Big upgrades for devs building with AI,” Microsoft CEO Satya Nadella wrote in an X post on Monday.
These models will not be available via ChatGPT and will be accessible only through the API.
However, GPT-4o-powered ChatGPT will gradually gain some of these new capabilities.
Microsoft has also announced that the GPT-4.1 models will be available to developers via the Azure OpenAI Service.
Compared to GPT-4o, the GPT-4.1 family of AI models is 26 per cent less expensive. GPT-4.1 Nano, in particular, will cost $0.10 for one million input tokens, $0.025 for cached input, and $0.40 for one million output tokens.
With this announcement, OpenAI has also confirmed the deprecation of GPT 4.5 Preview in the API, as GPT-4.1 offers similar or better performance at a lower cost and with reduced latency.
Bd-pratidin English/ Afia
