SemiAnalysis Launches GDPx; Opus 5.5 Leads at 12% — SkimNews
Get the Tech newsletter
Daily tech — startups, AI labs, chips, the launches that shape the next decade. Free.
- SemiAnalysis launched GDPx, an agentic capability benchmark covering 11 domains, with the listed fields running from RSI and hardware design to cyber, bio, finance, legal, quant trading, chip design and software.
- Opus 5.5 tops the GDPx leaderboard with a 12% score — the highest mark reported across the benchmark's 11 domains.
Why it matters: The material detail is the ceiling, not the crown: the best score on an 11-domain agentic suite is 12%. For teams evaluating models for chip design, cyber or RSI work, that 12% becomes the bar every competitor is measured against, and the wide domain spread from bio to quant trading shows the test is not narrow.
Ask SkimNews
