Does your work cost less on Opus 5.5 now that its tokens are 20% cheaper? My 3-prompt guide shows which jobs do before you move them.

Anthropic heard the complaints about Opus 5. Opus 5.5 and Sonnet 5.5 are both easier to steer, and both make even complex jobs cheaper to run.

TL;DR

  • Opus 5.5 and Sonnet 5.5 offer improved "ergonomics," making AI interactions easier to steer and control.
  • Enhanced usability allows users to provide direction, make corrections, and develop work without losing previous decisions.
  • Efficiency improvements enable users to accomplish more with their AI investments.
  • The focus on "model ergonomics" is key to AI helping build AI and accelerating development.
  • Anthropic's new releases suggest efficiency gains per task, potentially shaping compute availability.
  • Corrections that persist across connected problems enable larger, more ambitious assignments.
  • AI-assisted development by companies like Anthropic and OpenAI emphasizes faster releases while preserving user learning.
  • Repeat use and customer delegation are vital for the commercial success of AI companies.
  • A 3-prompt guide is suggested to test recurring work with new models.