OpenAI officially launched its latest GPT-5-Codex model API. This new model maintains the same pricing as its predecessor, GPT-5, allowing users to experience the power of this tool at a cost of $1.25 per million input tokens and $10 per million output tokens.

One of the major highlights of the GPT-5-Codex model is its exceptional context handling capability, with a context window of up to 400k. This means the model can process more complex information, significantly improving user efficiency in programming and other tasks. Additionally, the model's knowledge cutoff date is September 30, 2024, ensuring that users can access relatively up-to-date information and technical support.

OpenAI, artificial intelligence, AI

In terms of programming capabilities, GPT-5-Codex stands out particularly. Compared to previous models, it shows greater flexibility in responding to programming tasks, with "thinking" time ranging from a few seconds to seven hours, demonstrating its strong ability to solve complex programming problems. According to OpenAI's test results, GPT-5-Codex outperformed GPT-5 in the SWE-bench Verified benchmark for programming agents, especially in performance on large, mature codebase refactoring tasks, further solidifying its leadership position in the field of AI programming.

This advancement not only provides strong support for developers but also opens new doors for enterprises in software development, code optimization, and multiple other areas. By offering the tool through an API, OpenAI ensures its wide application across multiple development platforms, promoting the popularization and innovation of technology.

The launch of GPT-5-Codex means programming will become more efficient and intelligent. Developers can use this new tool to easily tackle various programming challenges, opening a new chapter in smart programming.