Claude Opus 5.5 fast mode leaves research preview
Opus 5.5 fast mode no longer needs an access request or a beta header. Anthropic prices it at $8 and $40 per million tokens.
Official source (opens in a new tab)
Anthropic's platform notes for 15 October 2026 say fast mode is out of research preview for Claude Opus 5.5. Requests no longer need the fast-mode-2026-02-01 beta header, though requests that still send it keep working, and you no longer need to request access. Start, Build, and Scale usage tiers get a default fast-mode rate limit. Fast mode for Claude Opus 5 and Claude Opus 4.8 stays in research preview. Opus 5.5 fast mode is also available on Claude in Microsoft Foundry when the deployment is hosted on Azure.
Anthropic's fast mode docs say this is the same model with a faster inference setup, up to 2.5 times the output tokens per second, turned on with speed set to "fast". For Claude Opus 5.5 that price is $8 per million input tokens and $40 per million output tokens.
Who it's for: developers already calling Claude Opus 5.5 who will pay more to stream tokens faster.
Fits: Best for Vibe Coding, on the API side. It does not change which coding tool ranks first.