OpenAI has expanded access to the massive context window available with GPT-5.6 Sol in Codex, allowing ChatGPT subscribers to work with much larger amounts of code and conversation history during a single coding session. OpenAI’s official specifications confirm that GPT-5.6 Sol supports a 1.05 million-token context window. A context window is essentially the amount of information an AI can keep in mind at one time, including prompts, source code, files, previous messages, and tool results.

OpenAI’s official specifications confirm that GPT-5.6 Sol supports a 1.05 million-token context window. A context window is essentially the amount of information an AI can keep in mind at one time, including prompts, source code, files, previous messages, and tool results.

According to OpenAI engineer Tibo, the full context length had previously been accessible in Codex through API-key usage but has now been enabled for ChatGPT accounts as well.

The larger window can be particularly useful when working with huge software projects. Codex can keep more files, previous changes, command outputs, and conversation history available before it has to compress older information.

That should make the feature useful for large codebase refactoring, complicated debugging, and long-running development sessions where losing earlier context can make an AI coding assistant less effective.

GPT-5.6 Sol is available to Codex users on supported ChatGPT plans, including Plus and Pro. OpenAI describes Sol as the GPT-5.6 model intended for its most demanding coding and reasoning tasks.

GPT-5.6 Sol 1M in Codex. This used to only work for API keys, but we just flipped the switch and works for usage through ChatGPT accounts now too.

The same warning applies, there is a reason the current context length is the default, we have tuned it to ~perfection. But you do… https://t.co/aYQfbY6e3L

— Tibo (@thsottiaux) August 17, 2026

The larger context window does not mean subscribers receive a larger usage allowance.

OpenAI says Codex usage depends on several factors, including the selected model, context size, reasoning, tool use and caching. Tokens are the small pieces of text and data the AI processes, so feeding Codex substantially more information can consume more of a user’s allowance.

For GPT-5.6 Sol, OpenAI currently estimates around 10 to 100 local Codex messages per five-hour period on Plus, although actual usage can vary considerably. Additional weekly limits can also apply. Pro plans provide higher allowances.

OpenAI itself recommends limiting unnecessary source material and context when users want their Codex allowance to last longer.

The million-token option therefore provides significantly more room when a difficult project genuinely needs it, but enabling such a large context for routine coding could cause available usage to disappear considerably faster.

Get the latest tech news, telecom insights, and product launches wherever you prefer.

Add ProPakistani to Preferred Sources and see more of our stories in Google Search and Top Stories.

Technology and Automotive Specialist covering the latest cars, smartphones, AI breakthroughs, and...