gpt-6VIDEO
三个 GPT-6 Astra 智能体协作跑分初测
在 Alem Hard 基准上,三个 Astra 智能体协作取得 25.2% 成绩,高于榜单当前最佳约 19%,但仅一个随机种子。
Kale-ab Tessera
@KaliTessera2026.09.17收录原推发布于 2026.09.16
Kale-ab Tessera @KaliTessera原推发布于 2026.09.16
Okay, first run of 3× GPT-6 Astra agents coordinating on Alem (Hard) is looking pretty interesting 👀 25.4% Base, 25.2% Coord. (current best Hard Coord. on the leaderboard is ~19%) Just one seed so far, so very preliminary.
x.com/kalitessera/status/2100243157946446294采集于 2026.09.17 10:17