Claude Opus 5.5 fast mode exits preview, goes GA

Anthropic has just rolled out Claude Opus 5.5 fast mode in general availability on the Claude Platform, dropping the waitlist and beta header requirements. The opt-in speed setting promises up to 2.5x faster output, with Azure availability now and Google Cloud support planned.

claude cover

TL;DR

  • Claude Opus 5.5 fast mode GA: On the Claude Platform; research preview ended; waitlist removed
  • Performance claim: Up to 2.5× higher output tokens/sec for supported models
  • Pricing (Opus 5.5 fast): $8/M input tokens; $40/M output tokens
  • Supported models: Opus 5.5, plus Opus 5 and Opus 4.8 at premium pricing
  • API usage: Opt in with speed: "fast"; beta header no longer required; header still accepted
  • Cloud availability: Microsoft Foundry on Azure; Google Cloud support planned later

Claude Opus 5.5 fast mode is now generally available on the Claude Platform, according to an announcement from the Claude Developers account quoted by Chubby. The service has left research preview, and its waitlist has been removed.

Claude Developers claims fast mode can deliver up to 2.5x higher output tokens per second for supported models. Opus 5.5 fast mode costs $8 per million input tokens and $40 per million output tokens.

Fast mode also supports Claude Opus 5 and Claude Opus 4.8 at premium pricing. Requests can opt in by setting speed: "fast". Opus 5.5 no longer requires the fast-mode-2026-02-01 beta header or a separate access request, although requests that continue sending the header will still work.

The accelerated mode is available through Microsoft Foundry, hosted on Azure. Support through Google Cloud is planned for a later date.

Fast mode in Claude Code remains unchanged and continues to use additional usage credits.

Source: X

Continue the conversation on Slack

Did this article spark your interest? Join our community of experts and enthusiasts to dive deeper, ask questions, and share your ideas.

Join our community