Anthropic has launched Claude Sonnet 5.5, the latest addition to its Claude 5.5 family. The new AI model arrives as the successor to Claude Sonnet 5 and focuses on three areas in particular: performance, speed and cost efficiency.
Anthropic says Sonnet 5.5 can complete the same work with fewer tokens than its predecessor. The company claims it is more than 30% faster at generating responses and can cost up to 30% less per task on typical workloads. Despite these improvements, its token pricing remains unchanged at the Sonnet 5 level.
The new model is aimed at everyday coding, agent-based tasks and knowledge work. It is also available across Anthropic’s own platform and major cloud services.
Claude Sonnet 5.5 Gets A Major Coding Boost
Coding is one of the biggest areas where Anthropic is positioning Sonnet 5.5 as an upgrade. On Terminal-Bench 4.0, an agentic coding evaluation, Sonnet 5.5 scored 70.6%, according to Anthropic. The company says this is a major jump over Sonnet 5 and also puts the new model ahead of Opus 5.5 on this particular evaluation.
The model is designed to understand a codebase before making changes. It can scope a task, make smaller edits and review its work. Anthropic says this makes it suitable for building features, fixing bugs and reviewing existing code.
Sonnet 5.5 also needs fewer tokens to complete tasks. This matters for developers running AI agents repeatedly because lower token usage can translate into lower overall costs.
Faster Responses And Lower Cost Per Task
Sonnet 5.5 improves on Sonnet 5 across benchmarks, in some cases dramatically.
It’s a faster, lower-cost complement to Claude Opus 5.5, strongest at well-scoped everyday tasks, fixing bugs, and creating polished documents, slides, and spreadsheets. pic.twitter.com/w8zB9WXxK7
— Claude (@claudeai) September 28, 2026Anthropic says Sonnet 5.5 generates outputs 30% or more faster than Sonnet 5. The company describes it as its fastest Sonnet model so far.
The speed improvement is not the only change. Anthropic says the model typically uses fewer tokens to complete the same work. As a result, the company estimates that Sonnet 5.5 can cost up to 30% less per task than Sonnet 5, even though the listed token prices have not changed.
That combination could be particularly useful for high-volume AI applications, coding agents and businesses running repeated workflows.
Claude Sonnet 5.5 Price And Availability
🤖 AI Benchmark Showdown!
Sonnet 5.5, Sonnet 5, Opus 5.5 & GPT-6 Sol compared across coding, reasoning, knowledge work and more. 📊
📈 Sonnet 5.5: 70.6% Terminal-Bench
🔥 Opus 5.5: 67.7% Humanity’s Last Exam#ClaudeSonnet #ClaudeAI #GPT6Sol #AI #Claude @claudeai pic.twitter.com/g0EzhMwQgy
— Techotales (@techotales) September 29, 2026Claude Sonnet 5.5 retains the same base pricing as Sonnet 5. Developers pay $2 per million input tokens and $10 per million output tokens.
Cache reads are priced at $0.20 per million tokens. Anthropic says prompt caching can provide up to 90% cost savings, while batch processing can reduce costs by up to 50% for eligible workloads.
The model is available through Claude.ai on the web, iOS and Android. Developers can also access it through the Claude Platform. Anthropic has made the model available through Amazon Web Services, Google Cloud and Microsoft Foundry.
| Claude Sonnet 5.5 | Details |
|---|---|
| Input Price | $2 per 1M tokens |
| Output Price | $10 per 1M tokens |
| Cache Reads | $0.20 per 1M tokens |
| Typical Cost Efficiency | Up to 30% lower per task |
| Speed | 30%+ faster than Sonnet 5 |
| API Model ID | claude-sonnet-5-5 |
| Availability | Claude.ai, Claude Platform, AWS, Google Cloud, Microsoft Foundry |
| Main Focus | Coding, Agents, Knowledge Work, Design |
Better Agents And Knowledge Work
Sonnet 5.5 is not limited to programming. Anthropic says the model has improved across knowledge-work tasks, including analysis, documents and presentations.
The company also highlights improvements in long-horizon agent tasks. Sonnet 5.5 can handle well-defined investigations, reviews and drafting jobs while using fewer tokens than the previous generation.
There are also improvements in visual and design work. The model can help create user interfaces, diagrams, user flows and presentation slides. Anthropic says its writing is clearer and more direct, which can reduce the amount of editing required afterward.
Claude Sonnet 5.5 Safety And Cyber Capabilities
Safety is another important part of the release. Anthropic says Sonnet 5.5 improved on or matched Sonnet 5 across most measures in its automated behavioural audit.
The company has also highlighted stronger cybersecurity capabilities. Because of those capabilities, Sonnet 5.5 is launching with cyber safeguards and fallbacks similar to those used with Anthropic’s more capable models. Its biology safeguards remain aligned with Sonnet 5.
Anthropic says these safeguards target a narrow set of high-risk requests. Routine software development and most life-sciences work are not affected.
Claude Sonnet 5.5 Vs Opus 5.5
Sonnet 5.5 is positioned as a faster and lower-cost complement to Claude Opus 5.5, rather than a direct replacement for it. Anthropic says Sonnet 5.5 can perform comparably to Opus 5.5 on some evaluations, particularly at higher effort levels. However, the company also notes that Opus 5.5 remains stronger for complex, open-ended work that requires sustained judgment.
This makes the two models serve different workloads. Sonnet 5.5 is aimed at tasks that need strong performance at scale, while Opus 5.5 remains positioned for more demanding work.
Conclusion
Claude Sonnet 5.5 brings a substantial performance upgrade without increasing the standard token price. Its biggest changes are faster responses, lower token usage and improved coding and agent capabilities.
Anthropic’s claim of up to 30% lower cost per task could be particularly important for developers and businesses running AI agents at scale. At the same time, its availability across Claude, the API and major cloud platforms makes the model accessible to both individual users and enterprise developers.
With Sonnet 5.5, Anthropic is essentially pushing more capability into its mid-range model while keeping the entry-level token pricing unchanged.
Follow Us: Facebook | X | Instagram | YouTube | Pinterest
