OpenAI has updated the default context window for its GPT-6.1 Sol model, increasing it from 272k to 500k tokens. This adjustment represents an approximate 80% expansion in capacity for users working within Codex and Work environments.
According to Thibault Sottiaux, the change aims to reduce the frequency of context compression triggers. The company notes that this 500k limit is currently considered the optimal balance between performance and user experience for complex tasks requiring long-term data retention.
Despite the update to the default setting, users retain the ability to manually adjust their context limits. The absolute maximum context length supported by the GPT-6.1 Sol model remains at 1M tokens.
Alongside these technical changes, OpenAI has updated its billing policy for the model. The company confirmed that it will no longer apply additional long-context fees when deducting usage from user subscriptions.
