[KongchangAI]
Unverified50% confidenceReleaseExact time

OpenAI announced that the API now officially offers an optional 12-hour ultra-long Prompt cache, while the cache for Codex and Work logged-in users' drawings is retained permanently.

1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/21/2026
First Seen
Valid until: 1/19/2027

Sources

DeepSeek Open-Sources Inference Framework, Shifting the AI Compute Landscape

Related Entities

OpenAIcompanyPrompt Cachingconcept

Related Claims

UnverifiedOn day 12 of OpenAI's 28-day consecutive release plan, OpenAI made context and request handling optimizations to Codex and ChatGPT Work, including reusing context across sub-agents, extending cache duration, and improving request routing to increase cache hit rates.73% similar

Cite This Claim

Stable URI
https://kongchang.com/claim/1081804
API
curl https://kongchang.com/api/v1/knowledge/claims/1081804
MCP
get_claim(id=1081804)