AInewsletter.semianalysis.comabout 2 hours ago
SemiAnalysis Launches Agentic Benchmark GDPx
16 min read
SemiAnalysis launches GDPx, a new benchmark measuring 11 domains including chip design and cybersecurity. Opus 5.5 tops the list with a 12% score, while Gemini 4 Carbon emerges as a strong contender. The tool partners with 14 specialized data providers to evaluate frontier models, addressing limitations in older benchmarks like GDPval.
Level
Hype check
Tap to vote and see what everyone thinks.
