0.5M 参数模型 VINCO 在卡坦岛 40 局中 36 局击败 Claude Opus 5.5 等模型

Rohan Paul· @rohanpaul_ai · X·2026-11-06 05:27· 50 分钟前AI 评分41

AI 导读

Distributed Cognition 的 0.5M 参数模型 VINCO 在 40 局卡坦岛对局中，有 36 局击败 Claude Opus 5.5、Claude Fable 5.1、Astra 和 Kimi K3。

正文 · 原文

Distributed Cognition’s VINCO, a 0.5M-parameter model beat Claude Opus 5.5, Claude Fable 5.1, Astra and Kimi K3 in 36 of 40 in the the board game Catan.

Catan tests a skill coding benchmarks don’t: picking good moves without seeing the opponent’s cards or knowing the next dice roll.

The company’s broader view is that superintelligence will come from many models like this collaborating, rather than one model covering everything. By its numbers, VINCO is about 10 million times smaller than the frontier LLMs it beat in testing.

引用distributed cognition@distcog

We're excited to unveil VINCO, our frontier AI model trained to play and win 1v1 Catan. We've secretly deployed it on the http://Colonist.io global leaderboard, where it has reached both the #1 and #2 positions, consistently beating every human world champion.

在 X 查看被引用的帖子

来源：Rohan Paul · x.com