In Codex, GPT-6.1 Sol uses a default context window of 500,000 tokens, according to OpenAI’s Codex models page, and the same default applies in ChatGPT Work.
As of November 9, 2026, OpenAI’s API model pages list a 1,050,000-token context window and 128,000 maximum output tokens for GPT-6.1 Sol, GPT-6 Astra, GPT-6.1 Terra and GPT-6 Luna, so the Codex default is smaller than the full API limit. Sol usage drawn from a ChatGPT subscription has no long-context surcharge in Codex. Other figures in circulation, such as 258K, do not match OpenAI’s current documentation.
In the Codex CLI, the model_context_window setting in config.toml sets the context window tokens available to the active model, and Codex compacts the conversation history automatically once it reaches a token threshold. OpenAI notes that sessions holding more context use more of a plan’s allowance per message.
Related: Codex usage limits on each plan, How much does Codex cost? and Codex CLI 0.167 power-user guide
Sources: OpenAI, Codex Models; OpenAI API docs, GPT-6.1 Sol; Codex configuration reference
Image: ChatGPT logo by OpenAI, public domain (PD shape), via Wikimedia Commons. Placed on a white background and resized. ChatGPT and Codex are trademarks of OpenAI. FreeAnswers is not affiliated with or endorsed by OpenAI.