OpenAI’s GPT-4.1 Sets New Standard with 1 Million Tokens, Empowering Developers
OpenAI has launched its GPT-4.1 series, featuring GPT-4.1, GPT-4.1 mini, and GPT-4.1 nano, accessible via API. These models replace the GPT-4.5 Preview, offering enhanced code generation, instruction adherence, and long-context processing with a context window of up to one million tokens. This advancement allows for handling complete code repositories, extensive documents, and intricate agent workflows within a single API call.
OpenAI has launched its GPT-4.1 series, featuring GPT-4.1, GPT-4.1 mini, and GPT-4.1 nano, accessible via API. These models replace the GPT-4.5 Preview, offering enhanced code generation, instruction adherence, and long-context processing with a context window of up to one million tokens. This advancement allows for handling complete code repositories, extensive documents, and intricate agent workflows within a single API call.
GPT-4.1 outperforms its predecessors in coding tasks, achieving a 54.6% score on the SWE-bench Verified benchmark, surpassing GPT-4o’s 33.2% and GPT-4.5’s 38%. It also demonstrates superior accuracy in code diffs, with 53% on Aider’s polyglot benchmark, more than double GPT-4o’s 18%. Instruction-following has also been refined, with GPT-4.1 reaching 38.3% accuracy on Scale’s MultiChallenge, compared to GPT-4o’s 27.8%.
The GPT-4.1 mini model reduces inference latency by nearly 50% and cost by 83% relative to GPT-4o while maintaining comparable performance. GPT-4.1 nano is optimized for low-latency tasks, achieving 80.1% on MMLU, 50.3% on GPQA, and 9.8% on Aider’s polyglot coding.
The GPT-4.1 series supports long-context use without added costs, following standard API pricing for token usage. This makes it suitable for applications involving large document retrieval, repository comprehension, or complete project editing without premium pricing tiers.
GPT-4.5 Preview will be discontinued on July 14. OpenAI recommends developers using GPT-4.5 or GPT-4o mini evaluate migration paths to GPT-4.1. ChatGPT users will continue to use GPT-4o, with ongoing integration of instruction-following improvements. The GPT-4.1 models are exclusively available through the API.
Notably, OpenAI is anticipating substantial revenue growth, projecting $12.7 billion this year and potentially $29.4 billion by 2026, driven mainly by the success of its AI subscription services like ChatGPT Pro. This optimistic forecast comes despite increasing competition from emerging AI companies, such as DeepSeek, Baidu, and Alibaba, particularly in China. OpenAI is also considering a shift to a for-profit model and is nearing the completion of a significant funding round.
If you want to read more news articles like this, visit DeFi Planet and follow us on Twitter , LinkedIn , Facebook , Instagram , and CoinMarketCap Community .
“Take control of your crypto portfolio with MARKETS PRO, DeFi Planet’s suite of analytics tools.”
Disclaimer: The content of this article solely reflects the author's opinion and does not represent the platform in any capacity. This article is not intended to serve as a reference for making investment decisions.
You may also like
Digital Commodity Capital Adds XRP to Its Portfolio, Bolstering Institutional Interest
BlockDAG Drops Price Pre-Reveal as Fartcoin Gains Traction

XRP Price Target Debated; SHIB Burns and Unstaked Gains

TRON Surpasses Resistance, Unstaked Promises High ROI

Trending news
MoreCrypto prices
More








