• Home
  • Technology
  • Gaming
  • Entertainment
  • World & Business
  • Science
  • Sports
  • AI
HomeTechnologyGamingEntertainmentWorld & BusinessScienceSportsAI
    • HomeTechnologyGamingEntertainmentWorld & BusinessScienceSportsAI
    • Home
    • Technology
    • Gaming
    • Entertainment
    • World & Business
    • Science
    • Sports
    • AI
    AI

    NVIDIA AI Skill Benchmarks Show Agent Gains

    Nous Research integrates the SkillEvaluator tool into skill installations.

    NA
    NR
    2 Sources, 42d ago, first seen 42d ago

    TLDR

    NVIDIA AI reported benchmarking over 300 verified skills on identical tasks and models. Agents using the skills showed gains of 41 points in correctness, 39 in effectiveness, and 35 in efficiency. Nous Research stated it now runs SkillEvaluator on skill installs to check for PII, leaked secrets, Unicode smuggling, licensing, and security issues. The company applied the tool to its own bundled skills and improved 11 of them.

    Combined views

    318.7K

    2 Sources, first seen 42d ago

    Combined views

    318.7K

    2 Sources, first seen 42d ago

    2.8K likes
    2.8K likes
    77 comments
    1.6K saves
    216 reposts

    Sentiment

    Positive——Negative

    Summary

    Not enough discussion yet.

    No sentiment analysis available yet.

    77 comments
    1.6K saves
    216 reposts
    Today's Rank

    —

    Not ranked yet

    Today's Rank

    —

    Not ranked yet

    Sentiment

    Positive——Negative

    Summary

    Not enough discussion yet.

    No sentiment analysis available yet.

    2 Sources

    @NVIDIAAIWe benchmarked 300+ NVIDIA verified skills to see how much they actually help agents on real tasks. Same task, same model, same setup. The only difference was whether the agent had the skill. Across the benchmarks, skills improved correctness by 41 points, effectiveness by 39, and efficiency by 35. SkillEvaluator is open source if you want to test your own skills before you ship them.
    @NousResearchHermes now leverages NVIDIA's SkillEvaluator on skill installs, checking for PII, leaked secrets, Unicode smuggling, licensing and security issues before you confirm. We pointed it at our own bundled skills first, and used what it found to improve 11 of them.

    2 Sources

    @NVIDIAAIWe benchmarked 300+ NVIDIA verified skills to see how much they actually help agents on real tasks. Same task, same model, same setup. The only difference was whether the agent had the skill. Across the benchmarks, skills improved correctness by 41 points, effectiveness by 39, and efficiency by 35. SkillEvaluator is open source if you want to test your own skills before you ship them.
    @NousResearchHermes now leverages NVIDIA's SkillEvaluator on skill installs, checking for PII, leaked secrets, Unicode smuggling, licensing and security issues before you confirm. We pointed it at our own bundled skills first, and used what it found to improve 11 of them.