OpenAI Enables 1-Million-Token Context for GPT-5.6 Sol in Codex via ChatGPT Accounts
OpenAI has expanded access to the 1-million-token context window for the GPT-5.6 Sol model in Codex, allowing ChatGPT subscription users (such as Plus and Pro) to use it directly instead of relying solely on API keys. This update allows developers to process massive codebases, perform complex debugging, and maintain long-running sessions directly within their subscription plans without incurring separate API usage costs. While the full context limit is approximately 1.05 million tokens, OpenAI engineers warn that enabling it will rapidly consume subscription quotas, and the default context length remains optimized for performance and cost.
## BACKGROUND
OpenAI Codex is an AI-driven coding agent designed to automate software engineering tasks like writing code and fixing bugs, accessible via terminal, desktop, and IDEs. GPT-5.6 Sol is a frontier model from OpenAI that supports a massive context window of up to 1 million tokens per request.