Automate Basics
New model · TechCrunch

Anthropic releases Sonnet 5.5 with a focus on faster work tasks

By Automate Basics, written with AI from TechCrunch's original

· 2 min read

Hands compare a printed document with a laptop draft in daylight.
Image: AI-generated illustration.

Anthropic has released Sonnet 5.5, a model it positions for everyday work such as creating office documents and coding. The company claims it responds faster than Sonnet 5 while using tokens more slowly.

What changed with Sonnet 5.5

Anthropic has released Sonnet 5.5, the latest version of its mid-range AI model. According to TechCrunch, Anthropic says the model is 30 percent faster than Sonnet 5 and uses tokens at a slower rate. The company presents it as an assistant for routine work, including writing office documents and coding.

Anthropic also reports that Sonnet 5.5 scored better than Opus 5.5 on its agentic coding benchmarks. That is a result for a specific kind of coding task, not a claim that Sonnet is stronger than Opus at everything. TechCrunch reports that Anthropic is applying the same cyber safeguards to Sonnet 5.5 that it uses for Fable and Opus because of the model's cyber capabilities.

Who can use Sonnet 5.5, and what does it cost?

TechCrunch reports that Anthropic has released Sonnet 5.5, but its report does not specify which plans include the model or how someone can access it. It also gives no price for using Sonnet 5.5.

Anthropic's claim about slower token use is worth separating from a price claim. Tokens are units used to process AI requests, so using fewer of them could matter for work that involves many requests. The source does not give rates or enough detail to calculate a bill, though. Teams considering a switch should check the access terms and pricing available to them rather than assume the model will lower their costs.

How do I try Sonnet 5.5 at work?

A short office document is a sensible first task for someone who has access to Sonnet 5.5, because Anthropic identifies document creation as one of its intended uses. Give the model material you are permitted to use, ask for a draft, then compare the result with the source material. Check whether it omitted important points, introduced unsupported claims or chose a tone that does not fit the audience.

For a team that uses coding agents, a familiar, bounded coding task offers another way to assess Anthropic's performance claims. Check the resulting code and the resources used before relying on it. TechCrunch does not report an independent hands-on test, so the company's speed and token claims should not replace a check against your own work.

Frequently asked questions

Is Sonnet 5.5 faster than Sonnet 5?

Anthropic claims Sonnet 5.5 is 30 percent faster than Sonnet 5 and uses tokens at a slower rate, according to TechCrunch. The report does not provide an independent test or say how those claims will translate to every workplace task.

Is Sonnet 5.5 cheaper to use?

TechCrunch describes Anthropic's claim that Sonnet 5.5 uses tokens more slowly, but does not give a price, plan details or a worked cost comparison. Teams should not treat lower token use as a guaranteed saving without checking the terms for their own access.

What work is Sonnet 5.5 meant to handle?

Anthropic positions Sonnet 5.5 for everyday tasks, including creating office documents and coding. The company also reports strong results on its agentic coding benchmarks. Those benchmark results concern a particular type of coding work, so they do not establish how well the model handles every office task.

Tools in this piece

Written with AI from TechCrunch's original and published after automatic checks: every figure here appears in the original, and no sentence is copied from it. The picture is AI-generated. The original is the authority.

Source: Anthropic releases Sonnet 5.5, which it calls a significantly cheaper, faster work partner, TechCrunch, 28 Sept 2026.