SemiAnalysis releases GDPx agent benchmark; best model scores 12%
11/02/2026 — 11/02, 14:18·2 sources (all)·2 reports (all)·In the 2026-11-02 briefing
Key takeaways
- SemiAnalysis released GDPx, a new agentic benchmark spanning 11 domains including chip design, cybersecurity, RSI, finance, law, quantitativ…
- the authors read this as a capability boundary rather than proof that agents are useless.
Background & analysis
On 2 November 2026, SemiAnalysis launched GDPx, a new benchmark for agentic capability. According to Techmeme, it spans 11 domains, including RSI, hardware design, cybersecurity, bio, finance, legal, quant trading, chip design, software engineering, materials science, and alignment. The headline result is striking: Opus 5.5 topped the ranking with a score of just 12%, suggesting that today's agents still complete only a small fraction of real professional work.
The same day, DEV Community · AI covered the release. That post notes it carries no sponsored or affiliate links and that all facts were checked against the sources listed at its end on 2 November 2026. It stresses that 12% is not a typo, nor is it a claim that AI agents are useless. Instead, it frames the figure as a statement about the current limits of agentic capability. The author also discloses that GDPx was not tested hands-on, and none of the models were run on its tasks.
As of now, the episode rests on these two reports, with no further developments. No independent reproductions of the score have appeared, and SemiAnalysis has not issued additional results beyond the initial ranking.
AI-generated from 2 reports · updated 3 hours ago
Latest turnSemiAnalysis released a new benchmark called GDPx on November 2, where the best frontier AI model scored just 12%. According to DEV Community, the article says this is not a claim that AI agents are useless, but a pointer to the boundaries of current agent capabilities. The authors did not test the benchmark hands-on.
- Heat index
- 182
- All sources
- 2
- All reports
- 2
- First seen
- 4 hours ago
Reports on this story headlines open the original
SemiAnalysis released a new benchmark called GDPx on November 2, where the best frontier AI model scored just 12%. According to DEV Community, the article says this is not a claim that AI agents are useless, but a pointer to the boundaries of current agent capabilities. The authors did not test the benchmark hands-on.
DEV Community · AIAI score 58SemiAnalysis has released GDPx, an agentic capability benchmark spanning 11 domains including chip design, cybersecurity, RSI, finance, law, quant trading, materials science and alignment. Opus 5.5 tops the leaderboard with just a 12% score, underscoring how far agents still are from completing real professional tasks.
TechmemeAI score 56
Other stories people are talking about
- Heat index 637RisingDeepSeek open-sources dsh-superagent coding agent prototype12 sources
- Heat index 386SurgeOpenAI, Google, Microsoft, Anthropic draft AI training preference signal in robots.txt6 sources
- Heat index 304OpenAI trims funding target; Anthropic may get BB rating at IPO9 sources
- Heat index 286NewOpenAI Signs Four Agreements in Abu Dhabi, Pushes Fundraising3 sources
- Heat index 266NewChatGPT and Codex to Share One Quota Starting in December3 sources
How is heat calculated?About the methodHide
Heat counts how many independent sources covered a story in the last 48 hours: one source counts once no matter how many posts it published, decaying with a 24-hour half-life. What ranks first is what many people are talking about.
This page aggregates public feeds. Headlines and summaries are machine-organized and remain the property of the original authors; verify important facts at the source. If you believe a headline or summary infringes your rights, email the contact address in the footer with the page URL and basis of your claim, and we will remove or replace it after review.
- Surge
- Discussion rising fast
- New
- First report within 6 hours
- Rising
- Still gathering discussion