Claude Sonnet 5 outperforms Opus 4.8 on GDPval
The release of Claude Sonnet 5 brings significant improvements, particularly beating the flagship Opus 4.8 model on the GDPval benchmark for real economic work. Additionally, it features a massive 1 million token context window.
The focus is shifting towards practical economic utility rather than sheer model size.
- –Sonnet 5's GDPval score (1618) surpasses Opus 4.8 (1615).
- –Opus is now considered a luxury rather than the default choice.
- –The 1M context window provides immense capacity for large-scale tasks.
DISCOVERED
1h ago
2026-06-30
PUBLISHED
2h ago
2026-06-30
RELEVANCE
AUTHOR
Truntr_