Grok 4.6 ranks third on the Artificial Analysis Healthcare and Medical Index

xAI’s Grok 4.6 has landed in third place on the Artificial Analysis Healthcare and Medical Index, sitting behind only Claude Opus 5 (max) and Claude Fable 5 (with fallback). For a model released around August 12, 2026, breaking into the top three of a specialized medical benchmark this quickly is a meaningful signal about where xAI is focusing its engineering effort. The ranking matters partly because of how the index is constructed. It is not a single-task leaderboard. Medical knowledge accounts for 35% of the score, agentic knowledge work for 25%, non-hallucination rates for 15%, reasoning for another 15%, and agentic customer interaction for the remaining 10. What the numbers say Grok 4.6 scored 61 on the broader Artificial Analysis Intelligence Index, up from the 56 that Grok 4.5 posted. It puts Grok 4.6 on equal footing with GPT-5.6 Sol on the Intelligence Index while outpacing it on the healthcare-specific ranking. Agentic capability accounts for a combined 35% of the total score when you add the knowledge work and customer interaction components together. Grok 4.6 is priced at $2 per million tokens for input and $6 per million tokens for output. A crowded and fast-moving fie...