Back to feed
News Story
机器之心
1 sources

Anthropic launches Claude Sonnet 5.5: over 30% faster, priced at half of Opus 5.5

Anthropic has officially released Claude Sonnet 5.5, the second model in the Claude 5.5 family, offering over 30% faster speed, up to 30% lower cost, and input, output, and cache write pricing at half that of Opus 5.5. The model significantly outperforms Sonnet 5 on coding, agentic tasks, and long-horizon knowledge work benchmarks, approaching Opus 5.5 in some evaluations, and is now available on AWS, Google Cloud, and Microsoft Azure.

SynthePulse Insight · AI deep readingMembers

Claude Sonnet 5.5: At Half the Price, Closing In on Opus — Anthropic Bets Agent Scale-Out on 'Good Enough'

Version 1 · 1 source

Anthropic releases Sonnet 5.5, the second model in the Claude 5.5 family, priced at roughly half of Opus 5.5 with 30%+ faster speed, closing in on the flagship across multiple benchmarks, and equipped for the first time on a Sonnet model with an anti-reasoning-extraction safety classifier.

  • Claude Sonnet 5.5 is officially released and live on all platforms including AWS, Google Cloud and Microsoft Azure, with the model identifier claude-sonnet-5-5.
  • Compared with the previous-generation Sonnet 5, it runs 30%+ faster and can cut costs by up to 30% on most work.
  • Input, output and cache-write pricing is half that of Opus 5.5, with cache reads also at $0.20 per million tokens.
Open section navigationA Release That Doesn't Push the Frontier

A Release That Doesn't Push the Frontier

Anthropic has officially released Claude Sonnet 5.5, the second model in the Claude 5.5 family, and it is already live on all platforms, including Amazon Web Services, Google Cloud and Microsoft Azure; developers can call it on the Claude Platform using the model identifier claude-sonnet-5-5.

Notably, Anthropic explicitly states that Sonnet 5.5 does not further push the frontier of model capability, so its alignment evaluation focuses on a targeted set of risks. This positioning shapes the narrative center of gravity of the entire release: not raising the capability ceiling, but spreading sufficiently strong capability at lower cost.

Price and Speed: Designed for High-Frequency Agent Tasks

According to a report by Machine Heart, compared with the previous-generation Claude Sonnet 5, Sonnet 5.5 runs 30%+ faster and can cut costs by up to 30% on most work. On pricing, its input, output and cache-write prices are all half those of Opus 5.5, with cache reads also at $0.20 per million tokens.

These numbers point to a clear product intent: better suited as the scaled default model for high-frequency Agent tasks. Sonnet 5.5 needs fewer tokens to complete each task than Sonnet 5, so running costs are lower, and output generation speed is also 30%+ faster.

It should be noted that the above speed and cost figures come from Anthropic's claims as relayed in the report; they are source claims and have not yet been independently verified by third parties.

Free for now

Read the full analysis

5 more sections of analysis, plus the full takeaway

Loading

Credibility boundary

The factual basis of this article comes from Machine Heart's report on Anthropic's release, and is second-hand relay. The release, launch platforms and model identifier are confirmable facts; the speed, cost, pricing, all benchmark scores, safety audits and protections are source claims, some of which (such as 'comparable performance at similar cost') come from Anthropic's own account and lack third-party verification. Anthropic has not disclosed complete comparison data between Sonnet 5.5 and Opus 5.5 across all benchmarks.

Sources for this version

  1. Claude Sonnet 5.5上线！一半价格，性能逼近Opus 5.5机器之心

Primary report

机器之心

Primary source