Benchmark Radar
AI BENCHMARK PROFILE

DirectorBench

General AIMultimodal Perception

DirectorBench evaluates long-form video generation across 5 dimensions (script, visual, audio, cross-modal, stability) using 80 metadata entries, 7 user profiles, and 40 checkpoint criteria.

Released
2026-05-28
Readiness
Paper only
Primary field
General AI

Why it matters

Current video benchmarks focus on short clips and aggregate scores, missing workflow failures and user preferences. DirectorBench provides diagnostic, profile-aware evaluation to guide improvements in long-form video generation.

Motivation

Long-form video generation is rapidly moving from short, single-scene synthesis toward minute-long, multi-shot creation with narrative structure, cinematic control, audio, and cross-modal synchronization.

Primary resources

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