Benchmark Radar™
Star —

PRISM-Bench

Explore score history ↗

Evaluates text-to-audio-video generation with an audio-centric diagnostic benchmark comprising 900 human-verified samples, factorizing audio evaluation by audio type and sound-source visibility, and assessing four perceptual dimensions via 35 fine-grained criteria using an MLLM-as-a-Judge protocol.

Released
2026-09-04
Evaluates
General AI, Vision & 3D, cs.MM
Openness
unknown
Importer
Claire Radar
Review status
ai-name-audit-deferred

PRISM-Bench paper

No reported scores are on record for this benchmark yet.

Which sources cite PRISM-Bench?

Related General AI, Vision & 3D, cs.MM benchmarks