Data as of Jul 25, 2026 · Based on 2,718,867 AI responses across 9,511 prompts · See how Parse measures this
The LLM Data Companyllmdata.com · Low evidenceThe LLM Data Company trains frontier language models for critical domains like medicine, law, and finance, prioritizing accuracy and resistance to sycophancy over generalist capabilities. Their Kos series of medical models, starting with Kos 1 Lite, achieves state-of-the-art performance on HealthBench Hard by handling ambiguity and pushing back on incorrect instructions.
Parse Score