SAN FRANCISCO — At its first developers conference, Code with Claude, AI company Anthropic today launched the latest versions of its large language models (LLMs): Claude Opus 4 and Claude Sonnet 4.
These updates promise significant advancements for developers and engineering teams, particularly in coding and long-context reasoning. In all coding-related benchmarks, both models easily outperform their predecessors, though in some other areas like visual reasoning, the old Sonnet model actually outperformed the new one.
While Sonnet 3.7…