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

    Adonis Singh Shares Astra Max Eyebench Leaderboard

    Engineer Adonis Singh posted a leaderboard screenshot comparing astra max and sol max models.

    BP
    PS
    DF
    7 Sources, 26d ago, first seen 26d ago

    TLDR

    Adonis Singh posted on X a screenshot of an eyebench-v3 leaderboard table. The image ranks models including gpt-6-astra max and gpt-6-sol max by visual acuity bars. Singh stated that the astra version delivers stronger results than the sol version while using fewer output tokens. The post comes from an account belonging to an 18-year-old engineer who works on LLM inference and contributes to the MCBench Minecraft AI benchmark.

    Combined views

    448.8K

    7 Sources, first seen 26d ago

    Combined views

    448.8K

    7 Sources, first seen 26d ago

    3.6K likes
    3.6K likes
    119 comments
    485 saves
    194 reposts

    Sentiment

    Positive——Negative

    Summary

    Not enough discussion yet.

    No sentiment analysis available yet.

    119 comments
    485 saves
    194 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

    7 Sources

    @adonis_singhAstra-max scores 95% on eyebench-v3. At half the cost (outputting ~3.8x less tokens) of Sol-max, it scores nearly double.
    @scaling01RT @adonis_singh: This is just not fair
    @steipeteCan’t remember last time we had such a large jump in capabilities.
    @charliermarshRT @adonis_singh: Astra-max scores 95% on eyebench-v3. At half the cost (outputting ~3.8x less tokens) of Sol-max, it scores nearly double…
    @BorisMPowerGPT-6 is in a class of its own! (Why not flip the X axis?)
    @DanielleFongtotally loopy if you ask me

    7 Sources

    @adonis_singhAstra-max scores 95% on eyebench-v3. At half the cost (outputting ~3.8x less tokens) of Sol-max, it scores nearly double.
    @scaling01RT @adonis_singh: This is just not fair
    @steipeteCan’t remember last time we had such a large jump in capabilities.
    @charliermarshRT @adonis_singh: Astra-max scores 95% on eyebench-v3. At half the cost (outputting ~3.8x less tokens) of Sol-max, it scores nearly double…
    @BorisMPowerGPT-6 is in a class of its own! (Why not flip the X axis?)
    @DanielleFongtotally loopy if you ask me