SemiAnalysis introduces GDPx to assess AI across professional domains

First reported by SemiAnalysis at · Updated · 2 sources

According to Unite.AI, GDPx spans 11 professional domains and its first version includes 206 tasks developed with 14 specialized data partners. The publication reports that Anthropic's Claude Opus 5.5 tops the debut leaderboard with a score of 12.0%, ahead of the other 12 models ranked.

  • SemiAnalysis says the benchmark covers fields including hardware design, cybersecurity, biology, finance and alignment.

Covered by 2 publishers within 4 hours of the first report: 1 news report and 1 analysis.

Reporting1

Unite.AI SemiAnalysis Launches GDPx Benchmark Spanning 11 Agentic Domains · Jonas Reeve, Cognitive AI & AGI, AI Research Agent

Analysis1

SemiAnalysis GDPx: Measuring Frontier Model Capabilities Across the Most Valuable Domains · Max Kan

Discussion

Hacker News 3 points, 0 comments

Related

Moonshot AI introduces its Kimi K3.1 Pro language model
Anthropic introduces Claude Duet for background work across connected apps
US tech stocks tumble as concerns over AI spending intensify
Replit reportedly moves most AI spending from Anthropic to OpenAI

Topics Anthropic