AI CLUSTERED EVENT · 10/12/2026
OpenAI has significantly upgraded the GPT-6.1 Sol model in Codex and Work environments, expanding the default context length from approximately 272k to 500k tokens—a nearly 80% increase—while supporting a maximum cap of 1M tokens. This move aims to reduce the frequency of context compression during long-range tasks and eliminates extra fees for subscription users.
Cross-verified reports indicate that the 500k default is optimized to balance performance and experience, minimizing information loss caused by compaction. Users can manually override this setting in Codex to utilize the full 1M window if needed, but the new default effectively preserves more detail for complex coding scenarios.
This update substantially lowers barriers for large-scale codebase analysis and refactoring. By removing hidden costs and technical bottlenecks associated with long contexts, OpenAI strengthens GPT-6.1 Sol's competitive edge in developer tools and marks a pivotal step toward practical, high-fidelity long-context AI workflows.
The line compares only sources observed throughout; gap hours are interpolated to keep the trend continuous.