Lovable, the AI-powered app builder used by millions, has switched its default model to Claude Sonnet 5, Anthropic's freshly released mid-tier model. The move comes on the same day Anthropic announced the model publicly, making Lovable one of the first major platforms to ship it to end users. According to Lovable co-founder Fabian Hedin, the upgrade was straightforward: the model gets more done for less, and it knows when to say no.
Opus-level muscle at Sonnet prices
Claude Sonnet 5 is built to be the most agentic Sonnet model yet. It can make plans, use tools like browsers and terminals, and run autonomously at a level that, just a few months ago, required larger and more expensive models. That's a meaningful shift for a mid-tier model.
Sonnet 5 narrows the gap to Opus 4.8 performance, but at lower prices. It's a substantial improvement over its predecessor, Sonnet 4.6, on important aspects of agentic performance like reasoning, tool use, coding, and knowledge work.
On concrete benchmarks, the gains are real:
- On one agentic coding benchmark, Sonnet 5 scores 63.2%, compared to Opus 4.8's 69.2% and Sonnet 4.6's 58.1%.
- On computer use tasks, Sonnet 5 neared the accuracy of Opus 4.8 at a significantly lower per-task cost.
- On a knowledge work benchmark, Sonnet 5 actually slightly outperforms Opus 4.8.
Why Lovable made the switch
Lovable's platform puts code generation and UI building in the hands of non-engineers, which means the model running underneath needs to be both capable and safe. According to Hedin, Claude Sonnet 5 refuses dangerous requests clearly and consistently. "At Lovable we aim to provide powerful tools to millions of developers. A model that knows when to say no is just as important as one that can build," he said.
Don't miss what's next in AI
Join 300,000+ engineers and researchers who get the signal, not the noise.
- Full access to in-depth AI research breakdowns
- Be the first to know what's trending before it hits mainstream
- Daily curated papers, repos, and industry moves