Sonar ran Anthropic's latest model through its LLM evaluation framework, benchmarking against Claude Opus 4.8, and found that bug density dropped 14%, vulnerability density dropped 20%, and blocker-level security issues fell 75% compared to Opus 4.8.
Published
Signal category
Research & Knowledge
Quote
“We ran Anthropic's latest model through Sonar's LLM evaluation framework, the same benchmark we use to test every major model. Bug density dropped 14%, vulnerability density dropped 20%, and BLOCKER-level security issues fell 75%.”
— Yassin Shabir|Sonar team
Company
Sonar
Trusted by 7M devs, Sonar is committed to enabling developers and organizations to build better code for better software
- Industry
- Software Development
- Location
- Vernier, CH
- Company size
- 1,020 employees
Sonar is the trust and verification layer for AI code, and the industry standard for automated code review for 17+ years. Sonar delivers deterministic, repeatable, and actionable code verification at scale by integrating code quality and code security into a single platform. The company analyzes more than 750 billion lines of code daily to ensure software is secure, reliable, and maintainable. Sonar is rooted in the open source community and is trusted by 7M+ developers globally, including teams at Snowflake, Booking.com, Deutsche Bank, AstraZeneca, and Ford Motor Company. To learn more about Sonar, please visit: www.sonar.com
Founded 2008