Skip to content
Neural Daily

ToolsNews

Claude Opus 5.5 fast mode drops its waitlist

Anthropic has removed the waitlist for Claude Opus 5.5 fast mode. Since October 15 it works on the Start, Build and Scale tiers without an access request or a beta header.

By: Mikola Máté Patrik · · 2 min read

Claude Platform Docs: Fast mode documentation page
Image: Anthropic · Screenshot · Screenshot of official page · Source

What changed

Fast mode runs the same Opus 5.5 model on a faster inference configuration: up to 2.5x higher output tokens per second with no change in capability. You opt in with speed: "fast" on the request (Claude docs).

Opus 5.5 no longer needs the fast-mode-2026-02-01 beta header and gets a default rate limit on the Start, Build and Scale tiers. Fast mode for Opus 5 and Opus 4.8 stays in research preview.

Pricing

Opus 5.5 fast mode costs $8 per million input tokens and $40 per million output tokens across the full context window. Prompt-caching and data-residency multipliers stack on top.

Caveats

  • Gains are in output tokens per second, not time to first token.
  • Switching between fast and standard speed invalidates the prompt cache.
  • Not available with the Batch API or a Priority Tier commitment.
  • Available on the Claude API and Azure-hosted Microsoft Foundry deployments, not on Claude Platform on AWS.

Who should use it

It pays off where streaming long answers is the bottleneck: interactive coding, agents and live interfaces. If most of your latency is before the first token, fast mode helps little. Check the usage.speed field in responses to confirm which speed actually ran.

    Produced with the help of AI, reviewed by an editor. How we make our articles

    Reviewed and edited by the Neural Daily desk. About

    Related articles