level: business
anthropic released sonnet 5.5, the newest version of its mid-tier model, positioning it as a faster and cheaper work partner for everyday tasks like coding and document creation. the company says the model is 30 percent faster than sonnet 5, which launched about three months ago, and burns tokens at a significantly slower rate. this speed advantage makes it suitable for agentic deployments where multiple agents run without exceeding cost limits.
benchmarks show sonnet 5.5 outperforming opus 5.5 on agentic coding, likely due to its ability to spawn multiple agents efficiently. anthropic also claims the model has cyber capabilities comparable to opus 5, making it the first sonnet model subject to the same cyber safeguards as fable and opus. the company did not disclose exact pricing or token costs, but emphasized lower operational costs compared to its predecessor. a new haiku model is expected in coming weeks.
this release follows a wave of model updates from major labs, including openai's enhanced sol and luna models and meta's new model for smart glasses. sonnet 5.5 sits below opus in anthropic's hierarchy but offers agility for cost-sensitive tasks. the focus on speed and cyber safeguards reflects growing demand for efficient agentic systems in enterprise settings. developers can access the model through anthropic's api and claude platform.
why it matters: faster and cheaper models like sonnet 5.5 reduce the cost of running ai agents for coding and office tasks, making automation more accessible to businesses.