GPT-6.1 Sol Default Context Jumps to 500K
OpenAI increased GPT-6.1 Sol’s default context window in Codex and ChatGPT Work from 272K to 500K tokens. Users can configure up to 1 million tokens, reducing compaction during long coding and professional-work sessions.
This is a meaningful quality-of-life upgrade for long-running coding agents, though more context does not automatically mean better retrieval or faster responses.
- –The change represents roughly an 80% increase in default working context.
- –Fewer compactions should help preserve requirements, decisions, and repository details across extended tasks.
- –Codex users can override the default up to 1 million tokens when projects demand it.
- –The API supports a roughly 1.05-million-token context, but prompts above 272K incur higher pricing.
- –Developers should still monitor latency, context quality, and usage consumption on very large sessions.
DISCOVERED
2h ago
2026-10-12
PUBLISHED
2h ago
2026-10-12
RELEVANCE
AUTHOR
WorldofAI