Claude Opus 5.5 Fast Mode (Generally Available)
A faster way to run Opus 5.5 through the Claude API, now open to all developers with no waitlist, at a premium price.
Details
- Runs the same Opus 5.5 model with a faster serving setup, so answers come out up to 2.5x faster with no change in quality
- The speed gain is in how quickly text streams out, not in how fast the first word appears
- Priced at $8 per million input tokens and $40 per million output tokens
- Turned on per request with a single setting; no access request or beta flag needed for Opus 5.5 any more
- Available on the Claude API, Claude Managed Agents, and Microsoft Foundry deployments hosted on Azure, with Google Cloud coming soon. Not available with the Batch API
- Fast mode inside Claude Code is unchanged and still draws on extra usage credits
Best for
Apps and agents where waiting on long Opus 5.5 responses slows people down.
Who benefits
Developers building on the Claude API or Managed Agents who already pay for Opus 5.5 and want faster output.
When to use it
Speed of the answer matters more than cost, for example live user-facing tools or tight agent loops.
Join the newsletter to get ahead with AI.
I share strategies, new modules, free guides, and tools to get you better at using AI in your work.
Leading a team? See team programs →