OpenAI Doubles GPT-6.1 Sol Context and Removes Surcharge

OpenAI has significantly expanded the capabilities of GPT-6.1 Sol models in Codex and Work services, increasing the standard context window to 500 thousand tokens.

Users can now work with large volumes of code and text without needing to compress them beforehand, which improves development efficiency.

An important change was the cancellation of additional fees for using long contexts within subscriptions, although separate rates remain for the API.

The limit is set by default at 500K tokens, but it can be increased to 1 million through configuration files if necessary.

To configure the maximum window in the local Codex CLI or IDE environment, you need to modify the model_context_window and model_auto_compact_token_limit parameters.

Detailed documentation on configuration is available on the official developer website.