RTK July 15, 2026
RTK vs Caveman: real token savings in AI agents
RTK and Caveman promise 60-90% token cuts in agents. I measure both with real data: JetBrains' 8.5% finding, a 614M-token replay and the output-vs-prose gap.
Read more
Back
Articles with tag: Tokens
3 articles found
AI Agents July 13, 2026
AI Token Savings: Cut Agent Costs up to 99%
Cut AI agent costs up to 99% with prompt caching, model routing, LLMLingua and RTK. A practical playbook from real production systems.
Read more
AI Agents June 20, 2026
Caveman: The Skill That Teaches AI Agents to Shut Up
Caveman teaches AI agents to talk like cavemen. Julius Brussee's viral skill claims 75% fewer tokens — real Reddit measurements and the brutal verdict.
Read more