Compare Claude 2 and Claude Opus 4, both from Anthropic, in 2 community votes, claude opus 4 wins 100% of head-to-head duels, context windows of 100K vs 200K, tested across 5 shared challenges.
Pick Claude Opus 4. In 2 blind votes, Claude Opus 4 wins 100% of the time. That's not luck.
Claude 2 is cheaper on both — 1.4× input, 2.3× output
Claude Opus 4 uses 52.1x more hedging
Compare Claude 2 and Claude Opus 4, both from Anthropic, in 2 community votes, claude opus 4 wins 100% of head-to-head duels, context windows of 100K vs 200K, tested across 5 shared challenges.
Pick Claude Opus 4. In 2 blind votes, Claude Opus 4 wins 100% of the time. That's not luck.
Claude 2 is cheaper on both — 1.4× input, 2.3× output
Claude Opus 4 uses 52.1x more hedging