引入GPT-4.1 AI模型家族

Openai推出了其最新的旗艦人工智能模型GPT-4.1家族，其中包括主要的GPT-4.1模型

OpenAI is rolling out a new family of flagship GPT models: GPT-4.1, Mini, and Nano.

Openai正在推出一個新的旗艦GPT型號：GPT-4.1，Mini和Nano。

The models, which are now available through OpenAI’s API, are designed to serve a range of developer needs, from handling large-scale codebases to delivering fast, low-latency responses for simpler tasks.

現在可以通過OpenAI的API獲得這些模型，旨在滿足一系列開發人員需求，從處理大型代碼庫到​​提供快速，低延遲響應以實現更簡單的任務。

The standout feature of the GPT-4.1 models is their ability to process up to 1 million tokens of context in a single call. This is a huge increase over previous models like GPT-4, which supported up to 128,000 tokens.

GPT-4.1模型的出色功能是它們在單個呼叫中最多可處理100萬個令牌的能力。這比以前的型號（如GPT-4）巨大增長，該模型支持多達128,000個令牌。

The expanded context window allows the models to work with entire code repositories, lengthy documents, and complex workflows without losing track of what’s relevant.

擴展的上下文窗口允許模型與整個代碼存儲庫，冗長的文檔和復雜的工作流程一起使用，而不會失去相關內容。

OpenAI says that GPT-4.1 shows a 21% improvement in coding abilities over GPT-4 and a 27% improvement over GPT-4.5. The models also use 26% less computing power than GPT-4.

OpenAI表示，GPT-4.1的編碼能力比GPT-4的提高了21％，比GPT-4.5提高了27％。與GPT-4相比，模型還使用了26％的計算能力。

The new family is particularly good at editing large files, following instructions, and understanding images, putting it ahead of previous models on several benchmarks.

新家庭特別擅長編輯大型文件，按照說明和理解圖像，將其領先於以前的模型。

With this release, OpenAI is phasing out the GPT-4.5 Preview, which will be turned off in the API on July 14, 2025. The company says that GPT-4.1 offers similar or better performance at a lower cost and with less latency.

有了此版本，OpenAI將逐步淘汰GPT-4.5預覽，該預覽將於2025年7月14日在API中關閉。該公司表示，GPT-4.1的性能以較低的成本和延遲較小。

The models are trained on data up to June 2024, making them the most up-to-date among OpenAI’s offerings.

截至2024年6月，這些模型對數據進行了培訓，使其成為Openai產品中最新的。

GPT-4.1, Mini, and Nano are available exclusively via the API and are not integrated into the ChatGPT interface. OpenAI has worked closely with the developer community to optimize these models for real-world coding and instruction-following tasks.

GPT-4.1，Mini和Nano可通過API獨家提供，並且不集成到ChatGpt接口中。 OpenAI與開發人員社區緊密合作，以優化這些模型，以實現實際編碼和指導遵守任務。

The Nano version is the fastest and most cost-effective, making it suitable for classification and autocompletion, while the full-size model excels at handling complex, multi-turn agent workflows.

納米版本是最快，最具成本效益的，使其適合分類和自動完成，而全尺寸型號則在處理複合物，多轉動代理工作流程方面表現出色。