GPT-6.1 Sol’s default context window in Codex and ChatGPT Work increased from 272K to 500K tokens, about 80%.

GPT-6.1 Sol’s default context window in Codex and ChatGPT Work increased from 272K to 500K tokens, about 80%

OpenAI’s Thibault Sottiaux says the larger default should reduce compactions during long tasks and help preserve context and nuance; he describes 500K as the tested default sweet spot for most users. Users can set a lower or higher limit, up to 1M tokens (announcement).

For Codex and Work subscribers, OpenAI says GPT-6.1 Sol usage drawn from a subscription has no long-context fee (pricing clarification). The claimed 2–3x reduction in compactions and faster task completion comes from a reposting user, not from OpenAI’s announcement; OpenAI states only that compactions should occur less often.

The sources do not name a specific rollout time beyond the announcement’s “Day 7” post, nor do they quantify the reduction in compactions.

OpenAI announcement · Subscription pricing clarification · Linked post