List your company

Medical Diagnosis on DDXPlus n=50

78Top-1 AccuracyBMBE + GPT-5.4-nano
38.4848.745969.26Apr 21, 2026

Evaluation results

MethodLinks
BMBE + GPT-5.4-nanoSystem Category=BMBE + sensor LLM, Sensor LLM=GPT-5.4-nano, Abstention Mechanism=DHS-optimal τ*2026.047890819688
BMBE + Gemini 3.1 Flash LiteSystem Category=BMBE + sensor LLM, Sensor LLM=Gemini 3.1 Flash Lite, Abstention Mechanism=DHS-optimal τ*2026.047490779686
BMBE + GPT-OSS-20BSystem Category=BMBE + sensor LLM, Sensor LLM=GPT-OSS-20B, Abstention Mechanism=DHS-optimal τ*2026.046178708878
SA GPT-5.4System Category=Standalone LLM doctors (frontier), Abstention Mechanism=self-determine abstention2026.046084647870
SA Gemini 3.1 ProSystem Category=Standalone LLM doctors (frontier), Abstention Mechanism=self-determine abstention2026.046064639676
BMBE + MiniMax M2.5System Category=BMBE + sensor LLM, Sensor LLM=MiniMax M2.5, Abstention Mechanism=DHS-optimal τ*2026.046078649476
BMBE + Gemma 4 31BSystem Category=BMBE + sensor LLM, Sensor LLM=Gemma 4 31B, Abstention Mechanism=DHS-optimal τ*2026.045878609674
SA Llama-4-MaverickSystem Category=Standalone LLM doctors (frontier), Abstention Mechanism=self-determine abstention2026.045672579270
BMBE + Llama-4-ScoutSystem Category=BMBE + sensor LLM, Sensor LLM=Llama-4-Scout, Abstention Mechanism=DHS-optimal τ*2026.045670619273
SA Qwen 3.6 PlusSystem Category=Standalone LLM doctors (frontier), Abstention Mechanism=self-determine abstention2026.044872519466
SA Kimi K2.5System Category=Standalone LLM doctors (frontier), Abstention Mechanism=self-determine abstention2026.044258547463
SA GPT-OSS-120BSystem Category=Standalone LLM doctors (frontier), Abstention Mechanism=self-determine abstention2026.044048419858