OpenAI Quietly Reduces Codex Context Window to 272k Tokens

OpenAI has quietly reduced the maximum context window length for Codex, drawing attention from the developer community. Context length is a critical metric when handling large codebases or complex workflows, and this adjustment will directly impact existing developer habits.

Key Details and Reactions

As discovered on Hacker News, OpenAI reduced the context window for GPT-5.6 Sol in the code tool's PR by 100,000 tokens, bringing it down from 372,000 to 272,000 tokens. The official release notes referred to this reduction as a "bug fix."

This change has sparked community discussions regarding LLM context strategies. @PawelHuryn pointed out that OpenAI's approach to shrinking context contrasts sharply with Anthropic's截然相反的策略. Meanwhile, @petrusenko_max warned that with weakened long-context capabilities, developers must re-evaluate prompt design and improve token efficiency when dealing with larger codebases or complex workflows. Programming paradigms that previously relied on ultra-long contexts will be directly impacted.

2026-07-20 ~ 2026-07-20 · 5 related posts

• [source] Codex Context Window Shrinks to 272k — luckokkkk · 2026-07-20
• OpenAI Reduces Codex Context Length — petrusenko_max · 2026-07-20
• [source] GPT vs Claude: Context Window Strategy Divergence — PawelHuryn · 2026-07-20
• Context Window Strategy Differences: OpenAI vs Anthropic — PawelHuryn · 2026-07-20

1 near-duplicate retellings: petrusenko_max