Leaked Benchmarks Show Opus 5 Performance Leap

Leaked benchmarks for Anthropic's unannounced Opus 5 show major leaps in research math and long-context agent capabilities, while reportedly halving the price compared to its predecessor.

2026-07-30 ~ 2026-07-30 · 2 related posts

• Leaked Opus 5 Benchmarks: Major Jumps in Research Math and Long-Context — echen · 2026-07-30
• Opus 5 benchmarks: #2 in research math and long-context agents, half the price — echen · 2026-07-30