- Claude Sonnet 5.5 is Anthropic's faster, cost-efficient successor to Sonnet 5 AI model
- The model costs 30% less per task and delivers outputs 30% faster than Sonnet 5
- Claude Sonnet 5.5 is available on AWS, Google Cloud, Microsoft Azure, and Claude Platform
Anthropic has launched Claude Sonnet 5.5 as the successor to its Sonnet 5 artificial intelligence model, claiming that the latest model is faster and more cost-efficient while delivering stronger performance in coding and other knowledge-intensive tasks.
The Dario Amodei-led AI company said Claude Sonnet 5.5 is priced at the same rates as Sonnet 5, while its internal testing showed that the new model costs 30% less per task than its predecessor.
Anthropic has also described it as its fastest model so far.
ALSO READ: GPT-6.1 Astra: OpenAI Scraps Release Of Next-Generation AI Model Over Safety Concerns
Claude Sonnet 5.5 Price And Availability
Claude Sonnet 5.5 is now available across major cloud platforms, including Amazon Web Services, Google Cloud and Microsoft Azure. Developers can also access the model through Anthropic's Claude Platform,
The model is priced at $2, or around Rs 192, per million input tokens and $10, or roughly Rs 960, per million output tokens. Cache reads cost $0.20, around Rs. 20, per million tokens, while cache writes are priced at $2.5, or approximately Rs. 240, per million tokens.
Despite retaining Sonnet 5's pricing, Anthropic said Sonnet 5.5 is more cost-efficient. According to the company's internal testing, it costs 30 per cent less per task than its predecessor. Anthropic also said the new model generates outputs 30 per cent faster than Sonnet.
Claude Sonnet 5.5 Coding Performance
Coding is one of the key areas where Anthropic says Sonnet 5.5 has made significant progress.
On Terminal-Bench 4.0's Agentic Coding test, Claude Sonnet 5.5 scored 70.6 per cent. This was higher than the 10.3 per cent score recorded by Sonnet 5 and the 66.4 per cent score of Opus 5.5.
The model also scored 46.2 per cent on FrontierCode 1.1. That was below Opus 5.5's 54.4 per cent, but higher than Sonnet 5's showing, bringing the model closer to OpenAI's GPT-6 Sol, which scored 49.3 per cent on the benchmark.
Sonnet 5.5 can also identify and fix bugs in code, the company said, in addition to its abilities for software development and other technical work.
Cyber Capabilities And Safety
Anthropic has also highlighted improvements in the model's cyber capabilities. Because of these capabilities, the company said Claude Sonnet 5.5 is being deployed with the same safeguards used for Opus 5.5.
ALSO READ: AMD to Buy Fei-Fei Li's World Labs AI Startup for $8.2 Billion
The company also said Sonnet 5.5 improved or matched Sonnet 5 on “most measures” of alignment, resistance to misuse and honesty during Anthropic's automated behavioural audit.
The newest model is also intended to work with a wider variety of knowledge tasks, and Anthropic is pitching Sonnet 5.5 as a faster and cheaper alternative for developers and companies that want powerful AI smarts without paying for the more expensive Opus 5.5.
Essential Business Intelligence, Sharp Market Insights, Practical Personal Finance Advice, Daily Fuel, Gold and Silver Prices and Latest Stories — On NDTV Profit.