20 large models take on Slay the Spire at A20 with only one clear, losses mostly blamed on being too cautious
主流大模型挑战最高难度《杀戮尖塔》,全都死于太过稳健……
Researchers used SpireBench to have 20 large models play through the same 10 new seeds at A20 difficulty in Slay the Spire; across 200 runs, only GPT-5.6 Sol cleared the game once. Compared with four top streamers, the AI fought only about 0.9-1.4 elite battles in Act 1 on average, versus about 2.6 for the streamers, and the AI chose to heal at 57%-75% of campfires, versus about 9% for the experts. Counting the first act alone, the best-performing models are already close to top human players, but long-term strategic ability still shows a clear gap.
The article traces the models' failures to excessive caution and sets them against per-run data from top streamers, making the models' specific weaknesses in long-term resource management easy to grasp.
Source: YYS (Youyanshe) · yystv.cn