Benchmark Radar
AI BENCHMARK PROFILE

VGA-BenchV2

General AIMultimodal PerceptionBestiVictoryLab

Evaluates video generation quality and aesthetic value with 1,016 prompts, 60,000 videos, and 36,000 task-level annotations.

Released
2026-08-26
Readiness
Inspectable
Primary field
General AI

Why it matters

Provides human-aligned evaluation and an evaluation-to-optimization pipeline, enabling reinforcement learning-based generator fine-tuning.

Motivation

We introduce VGA-BenchV2, an extended human-aligned benchmark and optimization framework for jointly evaluating and improving video generation quality and aesthetic value.

Primary resources

Benchmark Radar records only publicly supported details and links back to primary sources for verification.