AnalysisAI ModelsSeptember 19, 2026

Brood War Bench tests LLM agents in StarCraft: Brood War

Read original source →bw.swerdlow.dev

None of the tested models played beyond a beginner level; Codex Astra led, beating all other models consistently, while Grok models were not smart enough to play yet. Older models treated the RTS as turn-based and got destroyed while thinking; newer models were more cognizant of thinking costs.

1 source

More stories today

Open the live feed