RAG, Fine-Tuning & ContextElias Lumer et al.
Don't Break the Cache
Published "Don't Break the Cache" — measured prompt caching in the APIs of 3 providers: cost savings of 41–80%, though latency can increase depending on the placement strategy
Why it matters
The first measurement comparing the 3 providers under identical conditions since caching became a product
Source — Primary sources first; secondary reports are labelled “Press”
Paperarxiv.orghttps://arxiv.org/abs/2601.06007Added in Week 35, 2026 · See that week