List your company

Hallucination Detection on Earth Observation

90.94F1 ScoreGPT-5 nano
76.32880.121583.91587.7085Mar 20, 2026

Evaluation results

MethodLinks
GPT-5 nanoSize (B)=20*2026.0390.945.33
GPT OSSSize (B)=120A52026.0389.924.83
EVE-InstructSize (B)=242026.0384.73.5
Qwen3Size (B)=235-A222026.0384.42.17
MiniMax m2.5Size (B)=230A102026.0383.775.17
GPT-4.1Size (B)=1800*2026.0381.582.83
Mistral Medium 3.1Size (B)=200*2026.0376.894.17