Media benchmarks

Every model, same prompt

Real benchmark outputs matched to the models on our platform — browse every scenario by modality and category, then open a collection for the full comparison.

Curated collections

NormalizationDirections and units1 Invoice line1

ProsodyPunctuation3

FidelityFive nos1 Plain sentence1

CuesBracketed cue1 Nonverbal cue1

LanguageOther language3 Code switch1

LengthFull paragraph1 One word1

LiteralnessEmbedded instruction1

Text to AudioBenchmark16

Full paragraph

1 prompt16 models

Full paragraph · Length

View collection

Cookies keep FreeInd working. Optional analytics require your permission; ads for free accounts are served by Google, which shows its own consent message where required. Read our cookie policy.