Claude Opus 5.5 cuts flagship AI pricing by 20%

Anthropic artwork for the Claude Opus 5.5 model launch
Claude Opus 5.5 launch artwork. Image: Anthropic

Anthropic has launched Claude Opus 5.5, cutting the cost of its top Opus model while pushing performance closer to the company’s cheaper Fable tier. The release is available now across Claude products and the major cloud platforms, giving developers a less expensive option for demanding coding and agentic work.

In its official announcement, Anthropic says Opus 5.5 performs at roughly the level of Claude Fable 5.1 on most work, yet costs about 40% less to run in typical workloads. Standard API pricing is $4 per million input tokens and $20 per million output tokens, 20% below Opus 5. Cache reads cost $0.20 per million tokens.

A faster mode costs more

Developers who value speed can use Fast mode in Claude Code and the Claude Platform. Anthropic says it runs Opus 5.5 up to 2.5 times faster, but pricing rises to $8 per million input tokens and $40 per million output tokens. That makes standard mode the more economical choice when latency is not the main concern.

Don’t miss the best of The Mac Observer

Set us as a preferred source and our Apple reporting ranks higher in your Google Search results and Discover feed — one tap, no account changes.

Or get it by email

The model ID is claude-opus-5-5. It is available through Anthropic’s own platform as well as Amazon Web Services, Google Cloud, and Microsoft Azure. Anthropic also says zero data retention is available, although thinking mode cannot be disabled on this model.

The launch follows Anthropic’s steady expansion of its developer tools. The company recently raised Claude Code weekly limits, while its earlier Claude Opus 4.8 release focused on reliability and agentic workflows. Anthropic has also tested an Auto mode for Claude Code that reduces permission prompts.

For customers, the practical change is a lower entry price for Anthropic’s strongest broadly available Opus model, plus a separate speed option for time-sensitive tasks. The company’s performance comparisons are its own evaluations, so teams should still test Opus 5.5 against their actual workloads before switching production systems or estimating a new monthly bill.

Discussion

Join the discussionCommenting as a guest — your email is never published · Log in

Protected by Akismet — be kind, stay on topic.

This site uses Akismet to reduce spam. Learn how your comment data is processed.