한국어EN
RAG, Fine-Tuning & ContextOpenAI

API Prompt Caching

API Prompt Caching — 50% discount on cached input tokens

Why it matters

The last of the three companies to productize prefix caching

Source — Primary sources first; secondary reports are labelled “Press”

Officialopenai.comhttps://openai.com/index/api-prompt-caching/

Added in Week 35, 2026 · See that week

Same week

BIS creates DC VEU (Data Center Validated End User)GovernmentOpenAI releases the Realtime APIOfficialnano-graphrag v0.0.8 released: GraphRAG's core in about 1,100 linesOfficial

Nearby on the same axis

Contextual RetrievalOfficialJina AI presents Late ChunkingPaperAnthropic launches prompt caching in public betaOfficialContext Caching on DiskOfficial