Astra for Law leads in legal intelligence
Astra for Law 6.2 leads the Harvey LAB-AA benchmark with 20.8% full credit rate, achieves 95% fewer hallucinations than Claude Opus 5.5, and shows significant improvements in legal research accuracy and efficiency for ChatGPT Enterprise customers.