Skip to content
X · @elonmusk · X / Twitter

RT Polymarket: NEW: Grok 4.5 now scores highest on the SWE-Atlas-QnA benchmark, edging Claude Fable 5 & GPT-5.6 Sol.

RT PolymarketNEW: Grok 4.5 now scores highest on the SWE-Atlas-QnA benchmark, edging Claude Fable 5 & GPT-5.6 Sol.