ZhiKe AI

Apr 17, 2026 · Artificial Intelligence

Claude Opus 4.7 Boosts Programming Performance by 11% – Why Its ‘No’ Makes It More Reliable

Claude Opus 4.7 raises SWE‑bench Pro accuracy from 53.4% to 64.3% (a +11 pp jump), triples visual resolution, can refuse or verify dubious instructions, and keeps pricing unchanged while increasing token consumption, positioning it as a more reliable AI colleague despite a slight dip in long‑document search.