Unverified50% confidenceFactExact time
Developer Seth Wheeler replaced the meeting voting system originally driven by a small recurrent network with the open-weight model gpt-oss:20b in an Among Us simulator, and the result was that across six configurations, the large model never made the crew perform better.
1
Sources
50%
Confidence
Long-term
Relevance
10/16/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedGPT-6.1 Sol performs poorly in frontend and UI design, piling up low-quality UI and redundant copy in game generation tests.58% similarUnverifiedGPT-6.1 Sol's low cache price and million-scale context indicate it is aimed at agent and coding workflows rather than benchmark experiments.56% similarUnverifiedAI大模型（如DeepSeek、GPT-4）的技术本质是基于Transformer架构的大规模预训练语言模型，本身是无状态的56% similarUnverified任务委托机制的设计灵感来自丹尼尔·卡尼曼提出的双系统理论，GPT-Live 1 扮演系统1角色，GPT-5.5 扮演系统2角色56% similarUnverifiedOpenAI states that the work was driven by an internal model claimed to far surpass its newly released GPT-6 Astra, and that the team solving the Navier-Stokes problem involved approximately 10,000 concurrent agents.56% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1042910API
curl https://kongchang.com/api/v1/knowledge/claims/1042910MCP
get_claim(id=1042910)