Grok 4.6 Tops MedAgentBench Clinical Agent Benchmark

xAI's Grok 4.6 tops the MedAgentBench clinical agent benchmark with about 95.9% pass@1, beating GPT-5.6 Sol, while Grok 4.5 ranks third at roughly 93.4%.

2026-08-18 ~ 2026-08-19 · 2 related posts

• Grok 4.6 tops MedAgentBench, edging out GPT-5.6 Sol on agentic clinical EHR tasks — elonmusk · 2026-08-18

1 near-duplicate retellings: XFreeze