SemiAnalysis releases GDPx benchmark: AI agents score no higher than 12% on expert-level work across 11 fields
SemiAnalysis releases GDPx benchmark to test AI agents on expert-level tasks across 11 fields using work from 14 professional data partners. Claude Opus 5.5 leads with an average score of just 12%, while Gemini 4 Carbon scores 10.9%; GPT-6 Astra takes the lead in AI research and materials science tasks.
Continue looking at this topic
Hotly discussed viewsHere's what you might need: an open-source Grok Bot alternative written in Rust that comes pre-installed on your computer...OpportunitiesOne person runs a company in Hangzhou, using AI employees to take on development requests, while also discussing AI team building ↗Related contentWhat is Google Antigravity? Distinguish between the development platform and the Gemini model ↗Open-source projectthedotmack/claude-mem ↗Return to original text
A short introduction is retained here; the full text from the source website is not reposted. For specific wording and subsequent updates, please refer to the original article.
Open AIHOT original