List your company

Visual Question Answering on Colon-Bench Prompted

78.6AccuracyGemini 3 Pro
31.07243.41155.7568.089Mar 26, 2026

Evaluation results

MethodLinks
Gemini 3 Pro2026.0378.6
Gemini 3 ProPrompted=true2026.0378.6
Gemini 3 Flash2026.0376.6
Gemini 3 FlashPrompted=true2026.0376.6
Gemini 3.1 Flash Lite2026.0369.2
Gemini 3.1 F. Li.Prompted=true2026.0369.2
Seed 1.62026.0362.9
Seed 1.6Prompted=true2026.0362.9
GLM-4.6V2026.0355.7
GLM-4.6VPrompted=true2026.0355.7
Qwen3.5 397B2026.0349
Qwen3.5 397BPrompted=true2026.0349
Gemini 2.5 Flash Lite2026.0347.8
Gemini 2.5 F. Li.Prompted=true2026.0347.8
Qwen3-VL 235B2026.0346.7
Qwen3-VL 235BPrompted=true2026.0346.7
Molmo 2-8B2026.0346.1
Molmo 2-8BPrompted=true2026.0346.1
Qwen3.5 Plus2026.0344.3
Qwen3.5 PlusPrompted=true2026.0344.3
Qwen3-VL Plus2026.0341.2
Qwen3-VL 32B2026.0339.3
Qwen3-VL 32BPrompted=true2026.0339.3
Qwen-VL Max2026.0339.1
Qwen-VL MaxPrompted=true2026.0339.1
Seed 1.6 Flash2026.0338.1
Seed 1.6 FlashPrompted=true2026.0338.1
Qwen3-VL 8B2026.0332.9
Qwen3-VL 8BPrompted=true2026.0332.9