home.social

#gemini2_5 — Public Fediverse posts

Live and recent posts from across the Fediverse tagged #gemini2_5, aggregated by home.social.

  1. New benchmark shows Gemini 3 Pro outpaces Gemini 2.5 in trust, ethics and safety—69% vs 16%. The study, led by Phelim Bradley and Prolific, also pits DeepSeek V3 against the models, highlighting gaps in performance and reasoning. Dive into the full analysis for the numbers and implications. #Gemini3Pro #Gemini2_5 #DeepSeekV3 #TrustAndSafety

    🔗 aidailypost.com/news/gemini-3-

  2. New benchmark shows Gemini 3 Pro outpaces Gemini 2.5 in trust, ethics and safety—69% vs 16%. The study, led by Phelim Bradley and Prolific, also pits DeepSeek V3 against the models, highlighting gaps in performance and reasoning. Dive into the full analysis for the numbers and implications. #Gemini3Pro #Gemini2_5 #DeepSeekV3 #TrustAndSafety

    🔗 aidailypost.com/news/gemini-3-