All benchmarks

Media benchmarks

Compare generation quality across media models

Nonverbal cue

Why we chose this prompt: Parenthetical cues like (sighs) and (laughs) ask for a sound, not a word. This checks whether the model produces the nonverbal, skips it, or reads the cue text out loud.

Prompt

(sighs) Route 6 Espresso is open till nine. (laughs) The last train leaves at eleven fifty-six.

Sort by
  • Favicon for bytedance-seed

    provider default

    Cost
    --
    Generation time
    41.3s
    Correctness
    100%
  • Favicon for elevenlabs

    george

    Cost
    --
    Generation time
    2.2s
    Correctness
    100%
  • Favicon for elevenlabs

    george

    Cost
    --
    Generation time
    1.9s
    Correctness
    100%
  • Favicon for elevenlabs

    george

    Cost
    --
    Generation time
    2.0s
    Correctness
    100%
  • Favicon for elevenlabs

    george

    Cost
    --
    Generation time
    1.6s
    Correctness
    100%
  • Favicon for fish-audio

    provider default

    Cost
    --
    Generation time
    1.6s
    Correctness
    100%
  • Favicon for fish-audio

    provider default

    Cost
    $0.001
    Generation time
    1.4s
    Correctness
    100%
  • Favicon for fish-audio

    provider default

    Cost
    --
    Generation time
    1.2s
    Correctness
    100%
  • Favicon for google
    Google AI Studio

    Zephyr

    Cost
    $0.002
    Generation time
    5.5s
    Correctness
    100%
  • Favicon for google

    Zephyr

    Cost
    --
    Generation time
    3.9s
    Correctness
    100%
  • Favicon for minimax

    English_expressive_narrator

    Cost
    $0.009
    Generation time
    1.0s
    Correctness
    100%
  • Favicon for qwen

    longanlingxin

    Cost
    $0.002
    Generation time
    2.1s
    Correctness
    100%
  • Favicon for x-ai

    eve

    Cost
    --
    Generation time
    1.2s
    Correctness
    100%
  • Favicon for google

    Zephyr

    Cost
    $0.004
    Generation time
    3.9s
    Correctness
    95%
  • Favicon for microsoft

    cs-CZ-Grant:MAI-Voice-2.1-Flash

    Cost
    --
    Generation time
    2.3s
    Correctness
    95%
  • Favicon for minimax

    English_expressive_narrator

    Cost
    $0.006
    Generation time
    1.1s
    Correctness
    95%
  • Favicon for qwen

    loongjohn

    Cost
    $0.001
    Generation time
    1.6s
    Correctness
    84%
  • Favicon for microsoft

    cs-CZ-Grant:MAI-Voice-2.1

    Cost
    --
    Generation time
    2.6s
    Correctness
    80%
  • Favicon for deepgram

    flux-alexis-en

    Cost
    --
    Generation time
    7.1s
    Correctness
    79%
  • Favicon for assemblyai

    provider default

    Cost
    --
    Generation time
    1.2s
    Correctness
    75%
  • Favicon for elevenlabs

    george

    Cost
    --
    Generation time
    0.9s
    Correctness
    75%
  • Favicon for elevenlabs

    george

    Cost
    --
    Generation time
    1.3s
    Correctness
    75%
  • Favicon for elevenlabs

    george

    Cost
    --
    Generation time
    3.3s
    Correctness
    75%
  • Favicon for elevenlabs

    george

    Cost
    --
    Generation time
    2.3s
    Correctness
    75%
  • Favicon for elevenlabs

    george

    Cost
    --
    Generation time
    2.0s
    Correctness
    75%
  • Favicon for hexgrad

    af_alloy

    Cost
    --
    Generation time
    0.6s
    Correctness
    75%
  • Favicon for heygen

    nadine

    Cost
    --
    Generation time
    1.6s
    Correctness
    75%
  • Favicon for heygen

    nadine

    Cost
    --
    Generation time
    3.0s
    Correctness
    75%
  • Favicon for microsoft

    en-US-Harper:MAI-Voice-2

    Cost
    --
    Generation time
    1.7s
    Correctness
    75%
  • Favicon for microsoft

    en-US-Harper:MAI-Voice-2

    Cost
    --
    Generation time
    1.6s
    Correctness
    75%
  • Favicon for mistralai

    en_paul_sad

    Cost
    --
    Generation time
    4.8s
    Correctness
    75%
  • Favicon for sesame

    conversational_a

    Cost
    --
    Generation time
    122.7s
    Correctness
    71%
  • Favicon for fish-audio

    provider default

    Cost
    $0.001
    Generation time
    0.6s
    Correctness
    65%
  • Favicon for deepgram

    aura-2-thalia-en

    Cost
    $0.003
    Generation time
    3.2s
    Correctness
    62%
  • Favicon for canopylabs

    tara

    Cost
    $0.00067
    Generation time
    8.6s
    Correctness
    55%
  • Favicon for mistralai

    voice_3150

    Cost
    --
    Generation time
    19.1s
    Correctness
    --
  • Favicon for speechify

    geffen_32

    Cost
    --
    Generation time
    0.8s
    Correctness
    --