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

    Epoch AI Says Reasoning Models Double Progress Rate

    Jaime Sevilla of Epoch AI notes the Epoch Capabilities Index suggests doubled progress rate.

    3 Sources, 29d ago, first seen 29d ago

    TLDR

    Jaime Sevilla, co-founder and director of Epoch AI, posted that the Epoch Capabilities Index is hard to interpret yet indicates AI capabilities progress twice as fast since reasoning models were introduced. The statement appeared on X in a post from the EpochAIResearch account. Sevilla's bio in the post identifies him as a researcher focused on AI trends, scaling laws, and forecasting. No further details on the index movement or independent confirmation appear in the packet.

    Combined views

    —

    3 Sources, first seen 29d ago

    Combined views

    —

    3 Sources, first seen 29d ago

    — likes
    — likes
    — comments
    — saves
    — reposts

    Sentiment

    Positive——Negative

    Summary

    Not enough discussion yet.

    No sentiment analysis available yet.

    — comments
    — saves
    — reposts

    Sentiment

    Positive——Negative

    Summary

    Not enough discussion yet.

    No sentiment analysis available yet.

    Today's Rank

    —

    Not ranked yet

    Today's Rank

    —

    Not ranked yet

    3 Sources

    @JsevillamolThe ECI is a hard to interpret index, but this suggests that AI capabilities progress twice as fast since reasoning models were introduced.
    @GregHBurnhamAnother comment: had the pre-reasoning trend continued, we might just now be getting models around o3 / Gemini 2.5 Pro levels, or, top of band, Gemini 3 Pro / GPT-5.2. Esp if the former, AI feels downright slow! (Ofc hard to know what's intervened on in the counterfactual.)

    3 Sources

    @JsevillamolThe ECI is a hard to interpret index, but this suggests that AI capabilities progress twice as fast since reasoning models were introduced.
    @GregHBurnhamAnother comment: had the pre-reasoning trend continued, we might just now be getting models around o3 / Gemini 2.5 Pro levels, or, top of band, Gemini 3 Pro / GPT-5.2. Esp if the former, AI feels downright slow! (Ofc hard to know what's intervened on in the counterfactual.)