Codex Opens GPT-5.6 Sol's 1M-Token Context to ChatGPT Accounts

GPT-5.6 Sol's 1M-token context window in Codex now works for usage through ChatGPT accounts, not only API keys. That is the update shared by Tibo Sottiaux. (Tibo Sottiaux on X)

The same announcement points readers to a documented way to enable the larger context window and repeats an important caveat: Codex's current context length is the default because it has been tuned for performance and cost. (Tibo Sottiaux on X)

Key Takeaways

• GPT-5.6 Sol's 1M-token context option in Codex now works with ChatGPT accounts.
• The post says API keys previously were the only supported account type for this option.
• The default context length has not been replaced. The announcement says it was tuned for performance and cost.
• The source does not state plan names, regions, limits, or a general rollout schedule. This article does not infer them.

What changed

The access boundary is the main change. Sottiaux says GPT-5.6 Sol 1M in Codex previously worked only for API keys, then adds that the switch has been flipped for usage through ChatGPT accounts as well. (Tibo Sottiaux on X)

That makes the larger context window relevant to a different Codex credential path. It does not, by itself, establish that every ChatGPT plan, region, or Codex surface has the same access. The announcement does not provide those details, so they are intentionally left out here.

Why the default still matters

The announcement is unusually explicit that a larger context window is not automatically better for every task. Sottiaux says the current context length is the default because it was tuned for performance and cost, while the 1M-token option is available for users who need it. (Tibo Sottiaux on X; Tibo Sottiaux on X)

For a large repository or a long-running task, extra context can reduce the need to discard relevant material. It can also increase the amount of information the system has to process. Treat the larger window as a workload-specific setting rather than a universal upgrade.

What to verify before enabling it

The approved post links to instructions for enabling a 1M-token context window in Codex, but it does not reproduce the exact command or configuration in the announcement itself. Before using the option on a real project, verify the current instructions in the linked post and check:

• Whether the current Codex surface and account type expose the setting.
• How the larger window changes latency and usage for your repository.
• Whether the task actually needs more retained context than the default.
• Whether the workflow has a clear way to recover if a long context becomes unwieldy.

Those checks stay within the source's warning: the default remains the recommended balance for performance and cost unless a task benefits from more context.

What this announcement does not say

The source does not specify ChatGPT plan eligibility, supported regions, rate limits, pricing, exact token accounting, or whether the option is available in every Codex client. It also does not claim that a 1M-token context window improves every task. No such claims are added here.

Bottom line

Codex has extended GPT-5.6 Sol's 1M-token context option to usage through ChatGPT accounts, according to an OpenAI developer announcement. The larger window is an option, not a replacement for Codex's default context length, which the same source says was tuned for performance and cost. Use the linked instructions, then measure the setting on the kind of repository and task you actually run. (Tibo Sottiaux on X; Tibo Sottiaux on X)

Sources

• Tibo Sottiaux on X, GPT-5.6 Sol 1M in Codex for ChatGPT accounts (August 17, 2026): https://x.com/thsottiaux/status/2089143488696705077
• Tibo Sottiaux on X, instructions for enabling the 1M-token Codex context window (August 16, 2026): https://x.com/thsottiaux/status/2089082893804896524