OpenAI improves prompt caching for GPT-6
OpenAI says GPT-6 models hit the prompt cache more often by default, and discounts now apply to shared prefixes reused within 30 minutes. New tools include a caching dashboard, a diagnostics tool that explains cache misses, explicit cache breakpoints, and changing reasoning effort mid-conversation without losing the cache.
- Context
- 1.05M
- Max output
- 128K
- Input price
- $10.00
- Knowledge cutoff
- Apr 2026
Source: OpenAI news, checked Sep 28, 2026. Summarized from the provider’s announcement.