Fast mode for Claude Opus 5.5 leaves research preview

what changed

On 15 October 2026, Anthropic took fast mode for Claude Opus 5.5 out of research preview. You no longer need to request access.

Requests to Claude Opus 5.5 no longer need the fast-mode-2026-02-01 beta header. Requests that still send it keep working. Claude Opus 5.5 also now has a default fast mode rate limit in the Start, Build and Scale usage tiers.

Fast mode for Claude Opus 5 and Claude Opus 4.8 is still in research preview. For those models, keep sending the beta header.

The same day, fast mode for Claude Opus 5.5 became available on Claude in Microsoft Foundry, for deployments hosted on Azure.

why it matters

This changes a planning question. Fast mode was something you had to ask for and could not count on. For Claude Opus 5.5 it is now a normal option with a default rate limit, so you can design latency-sensitive paths around it.

Teams that mix Opus versions need care. One code path may now need the header for some models and not for others. Sending it to Claude Opus 5.5 is harmless, so the safest rule is to leave it in place.

The docs link to fast mode pricing and limits. We have not restated them here, so check the page before you plan cost.

what we would do

1. List the user-facing paths where Opus latency hurts, and test them on fast mode.
2. Check your tier’s default fast mode rate limit against your peak traffic.
3. Keep the header on for Claude Opus 5 and Claude Opus 4.8 calls.
4. Compare fast mode cost with your latency gain before you switch any path for good.

For model choice and cost trade-offs, see our assess service.