List your company

Medical Diagnosis on AgentClinic OOD original (test)

0.684Similarity (Sim)Aloe-Beta-70B
0.370960.452230.53350.61477May 8, 2026

Evaluation results

MethodLinks
Aloe-Beta-70BModel category=Medical open-source2026.050.68439.539.5
Aloe-Beta-70BModel / Variant=Aloe-Beta-70B2026.050.68439.539.5
MedExAgent-8BModel category=Medical open-source2026.050.67237.837.8
MedExAgent-8BVariant=full2026.050.67237.837.8
MedExAgent-8BVariant=w/o noise injection2026.050.66931.831.8
MedExAgent-8BVariant=w/o tool-match reward2026.050.65326.626.9
Qwen3-32BModel category=General open-source2026.050.6373136.8
Qwen2-32BModel / Variant=Qwen2-32B2026.050.6373136.8
HuatuoGPT-o1-70BModel category=Medical open-source, thinking mode=on2026.050.62523.824.9
HuatuoGPT-o1-70BModel / Variant=HuatuoGPT-o1-70B2026.050.62523.824.9
Qwen3-14BModel category=General open-source2026.050.61422.628.4
Qwen2-14BModel / Variant=Qwen2-14B2026.050.61422.628.4
MedExAgent-8BVariant=SFT-only2026.050.58928.628.9
Llama-3.1-8B-InstructModel category=General open-source2026.050.57522.529.8
Llama-3.1-8B-InstructModel / Variant=Llama-3.1-8B-Instruct2026.050.57522.529.8
Qwen3-8BModel category=General open-source2026.050.56515.321.9
Qwen2-8BModel / Variant=Qwen2-8B2026.050.56515.321.9
MedGemma-27B-text-itModel category=Medical open-source2026.050.54421.424.9
MedGemma-27B-text-itModel / Variant=MedGemma-27B-text-it2026.050.54421.424.9
MedExAgent-8BVariant=w/o cost penalty2026.050.38326.126.4