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

Open-ended generation on LLaVA-Bench In-the-Wild

109.3Score

Baseline

73.9483.1292.3101.48May 19, 2026
Updated 14d ago

Evaluation Results

MethodLinks
2026.05
109.3---
2026.05
106.9---
2026.05
106.2---
2026.05
104---
2026.05
100---
2026.05
99.8---
2026.05
99.1---
2026.05
85.5---
2026.05
84.7---
2026.05
84.6---
2026.05
84.1---
2026.05
81.1---
2026.05
80.4---
2026.05
75.3---
2025.03
-56.9471.8379.95
2025.03
-58.9474.7178.25
2025.03
-58.4471.8177.92
2025.03
-57.1769.481.22
2025.03
-54.4366.8974.81
2025.03
-59.4970.9381.63
2025.03
-61.678.2283.01
2025.03
-61.8877.4982.83
2025.03
-52.2969.7472.49
2025.03
-61.5977.9182.86
2025.03
-62.4677.2184.5