Skip to main content
ModelScale

Multimodal & Grounded benchmark

Facts-VLM leaderboard

Every model the catalog carries a published Facts-VLM value for, ranked by that value.

CategoryMultimodal & Grounded
MeasureEvidence-linked multimodal factuality
TasksGrounded factuality tasks
DifficultyGrounded multimodal factuality
Published byGLM-5V-Turbo

A grounded multimodal factuality benchmark for evidence-linked answer correctness.

No model in the catalog has a published Facts-VLM score.

The benchmark is defined by GLM-5V-Turbo, but the catalog carries no value for it yet. An absent value is shown as absent here rather than as a zero.

All leaderboards