Public Evaluations Audited
The story so far
Prime Intellect Blog found issues in AutomationBench and OfficeQA Pro v2 after auditing public evaluations of LLMs, leading to improvements that resulted in Claude Opus 5.5 achieving a highest score of 92.15%. The audit revealed problems in the existing benchmarks, prompting adjustments that enhanced performance metrics. The findings were shared by Prime Intellect Blog, focusing on the improvements made to the evaluation tools and their impact on LLM scoring.
AI-written from the reports · Updated 15 days ago
Coverage timeline
Follow the coverage to see every side of the story.
- Prime Intellect BlogNOV 3, 2026/ResearchLearnings from Auditing Public Evaluations
Prime Intellect's blog audit of the public evaluation found issues in AutomationBench and OfficeQA Pro v2, with improvements leading to Claude Opus 5.5 achieving a highest score of 92.15%