🎯 Prompt caching

The items that name it 13

• [AINews] Google admits AI Mode hurts Search revenue; -15% $500B drop worst since 2004 IPO. AI Bear Market coming?
  article2026-10-31

  Cuts agent task cost; Agent Arena data from 26K sessions shows 3.6x higher cost without it.

• Claude Code regularly sends each background agent’s transcript to its model
  writeup2026-10-29

  About 98% of agent summary calls re-read the transcript from cache, which makes them cheaper. The roughly 2% on a cold cache carry over a third of the cost.

• OpenAI and Anthropic cut their LLM prices by up to 80% within a few months. Why?
  article2026-10-25It is written here as Prompt caching (cache read pricing).

  cache read pricing

• TokenMaxxing: How to Get the Most Out of Your AI Models
  article2026-10-17
• Asana cuts model costs 76x in browser tests with GPT-6.1 Sol
  article2026-10-09

  Asana extended caching from fixed instructions and tool definitions to the agent's browsing history; 89% of input came from cache at 5% of the uncached price, making each call about 3x cheaper.

• [AINews] Claude Haiku 5.5 — better than GPT-6 Luna at the same pricing
  article2026-10-08It is written here as Prompt caching (cache reads / cache writes).

  cache reads / cache writes

• A model guide for the GPT-6 family
  article2026-10-02
• [AINews] Claude Opus 5.5, the new default model for AINews — and everybody cuts prices 40-50%
  article2026-09-23
• Better prompt caching for GPT-6
  article2026-09-22
• Introducing GPT-6 Sol and Luna
  article2026-09-22
• JFrog: Why Model Routing Backfires and How to Build Agents That Don't Burn Your Budget
  article2026-07-23
• rikugan: LLM reverse-engineering agent for IDA Pro and Binary Ninja
  repo

  Used with the Anthropic models to reduce cost.

• santosomar/AI-agents-for-cybersecurity
  repo