OpenAI announced today the launch of three new language models GPT-4.1, which are available via OpenAI API. News are named GPT-4.1, GPT-4.1 mini a GPT-4.1 nano. Not yet available for regular users ChatGPT, but is mainly aimed at developers and companies that work with APIs. Model GPT-4.1 brings significant improvements in programming, instruction understanding, and working with long context. It can process up to a million tokens at a time, allowing for better understanding of long texts or large documents. According to OpenAI has GPT-4.1 21,4% better results in programming tests than the model GPT-4o and 26,6% better than previous GPT.
You might be interested in
GPT-4.1 mini offers performance comparable to GPT-4o, but with lower latency and up to 83% lower cost. The smallest version, GPT-4.1 nano, is the fastest and cheapest model, ideal for simpler tasks like classification or text completion. With the introduction of these models OpenAI also ends support for the model GPT-4.5 in API because GPT-4.1 offers similar features at more favorable terms.
The price models look like this:
- GPT-4.1: $2 per million input tokens and $8 per million output tokens
- GPT-4.1 mini: $0,40 per million input tokens and $1,60 per output tokens
- GPT-4.1 nano: $0,10 per million input tokens and $0,40 per output tokens
OpenAI confirmed that some of the improvements from GPT-4.1 has already been ported to ChatGPT version GPT-4o and other features will be added gradually. All new models have a knowledge limit of June 2024.
