Claude Sonnet 5 Is Now the Default for Every Claude User

Anthropic has made Claude Sonnet 5 the default model for everyone who uses Claude, free and paid alike, putting what it calls its most capable mid-tier system in front of the widest audience it has ever reached in a single move.

The company introduced the model on June 30 and switched it on as the default a day later. The pitch is straightforward: most people should not have to think about which model they are talking to, and now the one they land on by default is meant to be good enough that they rarely need to reach for anything heavier.

Detail	What Anthropic says
Introduced	June 30, switched on as the default a day later
Who gets it	Everyone who uses Claude, free and paid alike
Introductory API price	$2 per million input tokens and $10 per million output tokens, through August 31
Standard price after that	$3 per million input tokens and $15 per million output tokens
Coding benchmark	A score in the low sixties on SWE-bench Pro, close to the flagship Opus 4.8
In Claude Code	Now the default, with a one-million-token context window and the same promotional pricing

What can Claude Sonnet 5 actually do?

Sonnet 5 is being sold on autonomy rather than raw chat. Anthropic describes it as the most agentic Sonnet it has built, a model that can lay out a plan, then carry it out using tools such as web browsers and terminals, and keep working on its own through tasks that used to demand a bigger and pricier system.

That framing matters because "agentic" is where the money is right now. Anthropic's coding tool, Claude Code, has become the engine of its business, and a capable default model that can run multi-step jobs without hand-holding is exactly what that customer base wants. On the industry's SWE-bench Pro coding benchmark, Sonnet 5 posts a score in the low sixties, close enough to the flagship Opus 4.8 to make the price gap tempting.

How much does Claude Sonnet 5 cost?

Anthropic is leaning on that gap. Through August 31 the model runs at an introductory API rate of $2 per million input tokens and $10 per million output tokens. After that, standard pricing kicks in at $3 and $15. Either way it undercuts the top-tier Opus, and the company is clearly betting that "nearly as good for a fraction of the cost" wins more developers than "best at any price."

For Claude Code users the change is already baked in. The tool now defaults to Sonnet 5 with a one-million-token context window and the same promotional pricing, available to anyone on a recent version.

The quiet strategy

Making a strong model the default is a bigger deal than it sounds. Most users never touch a model picker; they take whatever they are given. By putting Sonnet 5 in that slot, Anthropic guarantees its newest system is the one shaping the everyday experience of millions of people, not a premium option a curious few opt into.

It is also a statement about where the frontier is heading. When a company is comfortable handing its mid-tier model to every free user and calling it agentic, the definition of "good enough" has moved up a rung.