Hacker news

  • Top
  • New
  • Past
  • Ask
  • Show
  • Jobs

Two stealth models just appeared on the leaderboard, both ahead of Fable 5.2 (https://contralabs.com)

139 points by cheptsov about 22 hours ago | 45 comments | View on ycombinator

gorgmah about 18 hours ago |

I don't understand, where are those models available in stealth mode? Usually people do this on openrouter but I couldn't find any model called "gestalt". Does anyone know or is this just black hat marketing for this contralabs platform, and the model does not even exist?

gajus about 18 hours ago |

To clarify, it is not two models. Its a single model (Gestalt). The other bar on the benchmark is a human created dataset. Suffice to say we didn't expect anyone to beat it, but here we are. And no, it is not Contra model.

barbacoa about 16 hours ago |

It is possible that anthropic/open ai may no longer release their best models due to their massive size and cost of running, instead releasing distilled version for public use.

alex7o about 19 hours ago |

Glm 5.5 I bet, but still too early to tell

barrenko about 19 hours ago |

This Thanksgiving period will be a cambrian explosion of capable models.

irthomasthomas about 19 hours ago |

Deepseek V5 trained on Huawei Ascend?

curiousgeorge10 about 18 hours ago |

Very curious to find out which model that is :O

algoth1 about 15 hours ago |

The name gestalt indicates it’s probably a router model like paretto by unbiased

lyw6 about 17 hours ago |

interesting! any model performing better than human created outputs is very unexpected. i wonder how fast these models are gonna get better, thinking about the progress we made in a year on coding, design would be soon too

petesergeant about 18 hours ago |

This is a clever bit of content marketing

TheAtomic about 18 hours ago |

That web site is insufferable.

skinfaxi about 19 hours ago |

"Random AI lab makes up a new benchmark that puts at least one of its models ahead of Fable 5.2"

undefined about 18 hours ago |

undefined

cronin101 about 19 hours ago |

Saying this when none of the gallery examples show the new model's output is the ultimate "trust me, bro" move.

radzserg about 18 hours ago |

who is --John Galt-- Gestalt?

vonneumannstan about 18 hours ago |

Why would this not be GPT 6.2 Astra?

undefined about 16 hours ago |

undefined

ForHackernews about 18 hours ago |

"on the leaderboard [that we created so we could lead it]"

akutlay about 19 hours ago |

It is interesting that the Kimi model gets a very high “slop score” but “professional critiques” gave it a pretty high score. Do critiques have no taste or the “slop” is not slop anymore?

_s_a_m_ about 20 hours ago |

some obscure design benchmark

wangguode about 19 hours ago |

[flagged]

somelamer567 about 19 hours ago |

[flagged]