In May, Anthropic announced two new AI systems, class=”link ” href=”https://www.engadget.com/ai/anthropics-claude-opus-4-model-can-work-autonomously-for-nearly-a-full-workday-164526696.html” data-ylk=”slk:Opus 4 and Sonnet 4;elm:context_link;itc:0;sec:content-canvas”>Opus 4 and Sonnet 4. Now, less than six months later, the company is introducing Sonnet 4.5, and calling it the best coding model in the world to date. Anthropic’s basis for that claim is a selection of benchmarks where the new AI outperforms not only its predecessor but also the more expensive Opus 4.1 and competing systems, including Google’s class=”link ” href=”https://www.engadget.com/big-tech/google-rolls-out-its-most-powerful-gemini-model-yet-151205875.html” data-ylk=”slk:Gemini 2.5 Pro;elm:context_link;itc:0;sec:content-canvas”>Gemini 2.5 Pro and class=”link ” href=”https://www.engadget.com/ai/gpt-5-is-here-and-its-free-for-everyone-170001066.html?guccounter=1&guce_referrer=aHR0cHM6Ly93d3cuZ29vZ2xlLmNvbS8&guce_referrer_sig=AQAAACezLmWoxPMsxtV37FLypbESuxMgkZjaQK8vhYio-WDTgQI1K8lJKvj21T05s-piYZew5foyQ4Pxwe9PzxaIIFevvuj-Zc6QBdlt9AEeZgILLcbbHacXmXLVWxrY8S3svOH8V8ApZlhx_BmiAh605-Btyqw2zZ6vIRN3U4lY19fm” data-ylk=”slk:GPT-5 from OpenAI;elm:context_link;itc:0;sec:content-canvas”>GPT-5 from OpenAI. For instance, in OSWorld, a suite that tests AI models on real-world…
