Anthropic Launches Claude Opus 4.7 With 13% Coding Gains and New Effort Level
Anthropic released Claude Opus 4.7 on April 16, 2026, as its most capable publicly available AI model, succeeding Claude Opus 4.6. The update delivers a 13% improvement on the SWE-bench coding benchmark and resolves three times more production software tasks than its predecessor in internal evaluations. A new 'xhigh' effort level has been added to the API, designed to support long-running agentic workflows that require sustained multi-step reasoning. Vision capabilities have also been upgraded, with maximum image resolution rising from roughly 1.2 megapixels to 3.75 megapixels, pushing visual UI navigation accuracy from 57.7% to 79.5%. Anthropic's more powerful Claude Mythos Preview model remains restricted and is not part of the public API release.
This is an AI-generated summary. ShortSingh links to the original source for the complete article.
Discussion (0)
Log in to join the discussion and vote.
Log in