Claude Sonnet 5 may be the smarter default before you reach for Opus.
Anthropic's new Sonnet model targets professional coding, agentic work and reasoning at a lower price point than Opus, with temporary introductory API pricing through August.
Bit’s takeaway
What changed
Anthropic released Claude Sonnet 5 with introductory API pricing of US$2 per million input tokens and US$10 per million output tokens through 31 August 2026, before standard pricing of US$3 and US$15. It is positioned as the practical default for broad professional work.
Why it matters
A stronger flagship model is not automatically the better operating default. Sonnet may deliver enough quality with lower cost and latency for repeatable work, leaving Opus for tasks that fail a named quality test or genuinely need longer-horizon reasoning.
Who should care
- Teams choosing a default Claude API model
- People using Claude for routine coding, documents or agentic work
What to do
Start a representative workload on Sonnet 5, define the failure conditions that justify Opus, and track how often escalation actually changes the accepted result.
The human take
Tools change fast. Your judgment matters more.
This looks worth adopting, but prove it on one real task before you rely on it. You still own the judgment about where it fits.
Affected guidance
Put this to work
Sources and method