Glossary · Term

rate limit

← all terms

Definition

Plain language

A cap a website or service puts on how many requests you can make in a short window, so no single user can overwhelm it.

As stated in the literature

A server-side throttling control that rejects or delays requests exceeding a configured frequency, typically surfaced as an HTTP 429 response; a frequent obstacle for scraping and browsing agents, which sometimes try to route around it rather than back off.

Also called: rate-limiting, rate limiting, rate-limited, rate limits, rate-limit

Why it matters: It protects services from being overwhelmed, but it's also a common wall that browsing agents hit, and how they respond to it reveals whether they back off politely or try to cheat around it.

For example, a website might reject your requests with an error once you've made more than a hundred in a minute, forcing you to slow down.

Heard on the show

“An eighty-four percent failure rate — rate limits, timeouts — which roughly doubled the cost through wasted retries.”
Episode 191 — How One Researcher Beat GPT-5.2 and Gemini 3 by Judging Their Answers, Not Improving Them

Mentioned in 7 episodes

  1. 191
    How One Researcher Beat GPT-5.2 and Gemini 3 by Judging Their Answers, Not Improving Them
  2. 167
    How Teaching an AI to Predict, Not Act, Made It a Better Actor
  3. 119
    Beating Reinforcement Learning Without Ever Touching the Model's Weights
  4. 102
    How to Catch an AI Attack That No Single Conversation Reveals
  5. 080
    How a Two-Agent Trick Unlocked Large-Scale Training for Computer-Use Agents
  6. 061
    When Helpful Agents Go Sideways: A 404 Error, Campus Security, and Why Alignment Misses This
  7. 030
    Why Your AI Agent Won't Stop Working — and Each Model Falls for a Different Trap

Related terms