Anthropic launched Claude Sonnet 5.5 on Monday, September 28, 2026. The release spotlighted specifications and pricing that closely mirror those of the model introduced a week prior, Claude Opus 5.5, which debuted on September 22, 2026.
Sonnet 5.5 is priced at USD 2 per million input tokens and USD 10 per million output tokens—roughly half the cost of Claude Opus 5.5, which runs USD 4 per million input tokens and USD 20 per million output tokens. Furthermore, Sonnet 5.5 operates over 30% faster than its predecessor, Sonnet 5.
Price differences aside, Claude Sonnet 5.5 approaches the performance of Opus 5.5 across multiple key benchmarks. It achieved 70.6% on Terminal-Bench 4.0, outperforming the 66.4% score recorded by Opus 5.5. On an independent intelligence index, the new Sonnet model posted a score of 56 compared to 38 for the previous version. Even so, Anthropic maintains that Opus remains the superior choice for assignments requiring extended reasoning and deeper analysis.
Currently accessible via Amazon Web Services, Google Cloud, and Microsoft Azure under the identifier claude-sonnet-5-5, the model features zero data retention. Anthropic has also indicated that Claude Haikit 5.5 will roll out in the weeks ahead.
Also Read: Claude AI Outage Hits Users, 1,600+ Downdetector Reports
Why Cost Matters for Everyday AI Work
Price reductions carry considerable weight when organizations scale AI across numerous operations. Developers can deploy Sonnet 5.5 for writing software, debugging, and code reviews, as well as generating spreadsheets, presentations, and documents. For corporate environments, these incremental savings multiply rapidly when processing thousands of daily API calls. Quicker turnaround times and reduced token consumption further assist in managing expenses.
The arrival of Claude Sonnet 5.5 does not render Opus 5.5 obsolete, as the two systems cater to distinct tiers of responsibility. Sonnet efficiently manages standard workloads at a reduced expense, whereas Opus stays valuable for more complex challenges. Ultimately, users gain a budget-friendly alternative capable of supporting a broad spectrum of routine tasks without forcing them to default to the pricier model.
