Independent benchmarks for Opus 5.5: first place and third on secure code

Artificial Analysis put Opus 5.5 first on its Intelligence Index with 58 points, ahead of GPT-6 Astra and Fable 5.1 at 53. Endor Labs ranked it only third for secure code with 33.5% SecPass versus 37.4% for Fable 5.1, after removing 51 confirmed memorized solutions.

• Opus 5.5 tops Artificial Analysis Intelligence Index with 58 points
• Average cost per task: Astra $3.26, Opus 5.5 $5.98, Fable 5.1 $7.63
• SecPass: Opus 5.5 at 33.5%, Fable 5.1 at 37.4%
• Endor Labs confirmed 51 memorized solutions for Opus 5.5