Share your thoughts, 1 month free Claude Pro on usSee more

Abnormal Identification on PET-Bench

64.23AI Score

Gemini-2.5-pro

Updated 1mo ago

Evaluation Results

Method	Links
Gemini-2.5-pro 2026.01		64.23
GPT-4o 2026.01		55.86
GPT-5 2026.01		55.49
MedGemma-4B 2026.01		53.91
Qwen3-VL-30B 2026.01		53.78
Shizhen2.5VL-7B 2026.01		53.14
Lingshu-32B 2026.01		52.99
Lingshu-7B 2026.01		52.51
MedGemma-27B 2026.01		52.14
InternVL3.5-8B 2026.01		51.96
Shizhen2.5VL-32B 2026.01		51.3
InternVL2.5-38B 2026.01		51.1
Qwen2.5-VL-7B 2026.01		51.09
InternVL2.5-8B 2026.01		50.9
Qwen2.5VL-72B 2026.01		50.64
InternVL3-8B 2026.01		50.6
InternVL3-78B 2026.01		49.17
Grok-4 2026.01		49.1
Claude-sonnet-4.5 2026.01		48.97