OpenAI has officially released the API for its new GPT-5-Codex model. The pricing for this new model is consistent with its predecessor, GPT-5, with a rate of $1.25 per million input tokens and $10 per million output tokens.
One of the most significant features of the GPT-5-Codex model is its exceptional context handling, boasting a context window of up to 400k tokens. This allows the model to process more complex information, dramatically improving efficiency for users in coding and other tasks. Additionally, the model’s knowledge cutoff is September 30, 2024, ensuring users have access to a wide range of up-to-date information and technical support.
In terms of coding capabilities, GPT-5-Codex is particularly outstanding. Compared to previous models, its response time for coding tasks is more flexible, with “thinking” times ranging from a few seconds to seven hours, which demonstrates its powerful ability to solve complex programming problems. According to OpenAI’s test results, GPT-5-Codex outperformed GPT-5 on the SWE-bench Verified benchmark, especially in large-scale, mature code refactoring tasks, further solidifying its leadership in the field of AI programming.
This advancement not only provides strong support for developers but also opens new doors for companies in areas such as software development and code optimization. By offering the model via an API, OpenAI ensures that this tool can be widely applied across various development platforms, promoting technological adoption and innovation.
The launch of GPT-5-Codex means that coding will become more efficient and intelligent. Developers can leverage this new tool to easily tackle various coding challenges, starting a new chapter in intelligent programming.
© Copyright notes
The copyright of the article belongs to the author, please do not reprint without permission.
