AI Agent Revolution: How Anthropic Cut Token Usage by 98% with Code Execution
Last Updated on November 11, 2025 by Editorial Team
Author(s): AbhinayaPinreddy
Originally published on Towards AI.
The Problem That’s Been Quietly Killing AI Agents
Imagine hiring an assistant who needs to read a 500-page instruction manual before making a simple phone call. That’s essentially what’s been happening with AI agents — and it’s been draining budgets, slowing responses, and limiting what these agents can actually do.
The article discusses the significant advancements made by Anthropic in improving the efficiency of AI agents by drastically reducing token usage from 150,000 tokens to just 2,000 tokens, representing a 98.7% reduction. This transformation allows developers to create more complex agents without worrying about costs and latency, leveraging a new paradigm where agents write code instead of just calling tools. The article outlines the flaws of traditional tool-calling approaches, particularly in terms of token consumption, latency, and limitations in functionalities. By adopting code execution models such as Model Context Protocol (MCP), the AI agents can respond faster, maintain privacy better, and handle multiple tasks simultaneously, ultimately revolutionizing the way AI agents function.
Read the full blog for free on Medium.
Join thousands of data leaders on the AI newsletter. Join over 80,000 subscribers and keep up to date with the latest developments in AI. From research to projects and ideas. If you are building an AI startup, an AI-related product, or a service, we invite you to consider becoming a sponsor.
Published via Towards AI
Towards AI Academy
We Build Enterprise-Grade AI. We'll Teach You to Master It Too.
15 engineers. 100,000+ students. Towards AI Academy teaches what actually survives production.
Start free — no commitment:
→ 6-Day Agentic AI Engineering Email Guide — one practical lesson per day
→ Agents Architecture Cheatsheet — 3 years of architecture decisions in 6 pages
Our courses:
→ AI Engineering Certification — 90+ lessons from project selection to deployed product. The most comprehensive practical LLM course out there.
→ Agent Engineering Course — Hands on with production agent architectures, memory, routing, and eval frameworks — built from real enterprise engagements.
→ AI for Work — Understand, evaluate, and apply AI for complex work tasks.
Note: Article content contains the views of the contributing authors and not Towards AI.
Related posts
Recent Posts
