Handheld medium close-up of a chef tasting a sauce from a spoon, then looking up at the camera and saying: "Needs more lemon." Spoken audio with matching mouth shapes, kitchen clatter in the background, warm practical light.

Talking Head & Lip Sync · default settings, one run, 5 s, 720p, 16:9 · Updated September 20, 2026

Not run yet
MiniMax FastH3 V2
Not run yet
P-Video 2 Pro
Not run yet
Wan 3.0
Not run yet
Wan 3.0 Prime
FastH3 V2
4.6videomodels.ai score
  • Prompt adherence5.0
  • Motion & consistency4.5
  • Camera & framing4.0
  • Quality & detail4.5
  • Look & lighting4.5
  • Realism & physics4.5
  • Text rendering—
  • Audio5.0

The complete taste, assess, and spoken verdict sequence plays out cleanly.

Seedance 2.5
2.7videomodels.ai score
  • Prompt adherence3.0
  • Motion & consistency2.5
  • Camera & framing2.0
  • Quality & detail3.0
  • Look & lighting3.5
  • Realism & physics2.5
  • Text rendering—
  • Audio—

Chef tastes and turns to camera, but pronounced hat distortion and no clear speech hurt the result.

Seedance 2.0
3.7videomodels.ai score
  • Prompt adherence3.5
  • Motion & consistency4.0
  • Camera & framing3.0
  • Quality & detail4.0
  • Look & lighting4.0
  • Realism & physics4.0
  • Text rendering—
  • Audio—

Tasting and camera look are present, but the spoken delivery is only weakly conveyed.

Seedance 2.0 Fast
3.6videomodels.ai score
  • Prompt adherence3.5
  • Motion & consistency3.5
  • Camera & framing3.0
  • Quality & detail4.0
  • Look & lighting4.0
  • Realism & physics4.0
  • Text rendering—
  • Audio—

Convincing kitchen tasting setup, though the final spoken-to-camera beat remains implied.

Kling 3.0 Turbo
3.6videomodels.ai score
  • Prompt adherence3.0
  • Motion & consistency4.0
  • Camera & framing2.0
  • Quality & detail4.5
  • Look & lighting4.5
  • Realism & physics4.0
  • Text rendering—
  • Audio—

Strong visual sequence, but the requested speech is absent.

Veo 3.1
3.1videomodels.ai score
  • Prompt adherence2.5
  • Motion & consistency3.0
  • Camera & framing3.0
  • Quality & detail3.5
  • Look & lighting4.0
  • Realism & physics3.0
  • Text rendering—
  • Audio—

Strong warm kitchen setup and tasting action, but the direct-to-camera spoken conclusion is missing.

Veo 3.1 Fast
3.1videomodels.ai score
  • Prompt adherence2.5
  • Motion & consistency3.0
  • Camera & framing3.0
  • Quality & detail3.5
  • Look & lighting4.0
  • Realism & physics3.0
  • Text rendering—
  • Audio—

The tasting shot looks appealing, but he never clearly addresses camera or delivers the required line.

Happy Horse 1.1
3.6videomodels.ai score
  • Prompt adherence3.0
  • Motion & consistency4.0
  • Camera & framing2.0
  • Quality & detail4.5
  • Look & lighting4.5
  • Realism & physics4.0
  • Text rendering—
  • Audio—

Strong tasting sequence, but the requested spoken line is absent.

MiniMax H3
3.5videomodels.ai score
  • Prompt adherence3.5
  • Motion & consistency3.5
  • Camera & framing3.0
  • Quality & detail3.5
  • Look & lighting4.0
  • Realism & physics3.5
  • Text rendering—
  • Audio—

The spoken delivery is only weakly visible and the framing is too wide.

Gemini Omni Flash
3.6videomodels.ai score
  • Prompt adherence3.0
  • Motion & consistency3.5
  • Camera & framing3.5
  • Quality & detail4.0
  • Look & lighting4.5
  • Realism & physics4.0
  • Text rendering—
  • Audio—

Attractive kitchen close-up, though slight jitter and absent spoken audio hurt compliance.

Wan 2.7
3.4videomodels.ai score
  • Prompt adherence4.0
  • Motion & consistency2.5
  • Camera & framing3.0
  • Quality & detail3.5
  • Look & lighting4.0
  • Realism & physics2.5
  • Text rendering—
  • Audio4.5

The spoken line is present, but visible facial warping weakens the take.

Grok Imagine Video
3.7videomodels.ai score
  • Prompt adherence3.0
  • Motion & consistency4.0
  • Camera & framing4.0
  • Quality & detail4.0
  • Look & lighting4.0
  • Realism & physics4.0
  • Text rendering—
  • Audio—

The chef clearly tastes then addresses camera, but the requested dialogue has no audio.

P-Video 2
4.0videomodels.ai score
  • Prompt adherence4.0
  • Motion & consistency4.0
  • Camera & framing4.0
  • Quality & detail4.0
  • Look & lighting4.0
  • Realism & physics3.5
  • Text rendering—
  • Audio—

Clear prompt sequence and warm kitchen look, with slight eye-line inconsistency.

Runway Gen-4.5
3.7videomodels.ai score
  • Prompt adherence3.5
  • Motion & consistency3.5
  • Camera & framing3.5
  • Quality & detail4.5
  • Look & lighting4.0
  • Realism & physics4.0
  • Text rendering—
  • Audio—

Polished kitchen scene, but the requested spoken delivery is not convincingly visible.

Hailuo 2.3
3.4videomodels.ai score
  • Prompt adherence2.5
  • Motion & consistency3.5
  • Camera & framing3.5
  • Quality & detail4.0
  • Look & lighting4.0
  • Realism & physics3.5
  • Text rendering—
  • Audio—

The tasting reads cleanly, but the spoken line is not completed and no speech audio exists.

Kling 2.6
3.4videomodels.ai score
  • Prompt adherence3.0
  • Motion & consistency4.0
  • Camera & framing2.0
  • Quality & detail4.0
  • Look & lighting4.0
  • Realism & physics4.0
  • Text rendering—
  • Audio—

The tasting and reaction work, but requested dialogue is absent.

FLUX 3
3.8videomodels.ai score
  • Prompt adherence3.0
  • Motion & consistency4.0
  • Camera & framing4.0
  • Quality & detail4.0
  • Look & lighting4.5
  • Realism & physics4.0
  • Text rendering—
  • Audio—

Strong warm kitchen staging and complete visible action, but no required speech track.

PixVerse V5
3.7videomodels.ai score
  • Prompt adherence3.5
  • Motion & consistency4.0
  • Camera & framing3.5
  • Quality & detail4.0
  • Look & lighting4.0
  • Realism & physics3.5
  • Text rendering—
  • Audio—

The taste and camera look land, but spoken delivery remains unclear.

Seedance 1.5 Pro
4.4videomodels.ai score
  • Prompt adherence4.5
  • Motion & consistency4.5
  • Camera & framing4.0
  • Quality & detail4.5
  • Look & lighting4.5
  • Realism & physics4.5
  • Text rendering—
  • Audio—

A polished, coherent tasting-to-address-camera progression with only subtle generated-video drift.

FastH3
3.4videomodels.ai score
  • Prompt adherence2.5
  • Motion & consistency3.5
  • Camera & framing3.5
  • Quality & detail4.0
  • Look & lighting4.0
  • Realism & physics4.0
  • Text rendering—
  • Audio—

Convincing chef and tasting action, but the clip only begins the address and has no speech audio.

Cosmos 3 Super
3.2videomodels.ai score
  • Prompt adherence2.5
  • Motion & consistency3.5
  • Camera & framing3.5
  • Quality & detail3.5
  • Look & lighting3.5
  • Realism & physics3.5
  • Text rendering—
  • Audio—

The visible tasting progression works, but the subject is not a chef and speech is absent.

Lance
3.4videomodels.ai score
  • Prompt adherence3.0
  • Motion & consistency4.0
  • Camera & framing2.0
  • Quality & detail4.0
  • Look & lighting4.0
  • Realism & physics3.5
  • Text rendering—
  • Audio—

Good action staging, but the dialogue requirement is missing.

Wan T2V
3.4videomodels.ai score
  • Prompt adherence3.0
  • Motion & consistency4.0
  • Camera & framing3.5
  • Quality & detail4.0
  • Look & lighting3.5
  • Realism & physics4.0
  • Text rendering—
  • Audio2.0

The tasting sequence is clear, but there is no speech track.

HunyuanVideo 1.5
3.6videomodels.ai score
  • Prompt adherence3.0
  • Motion & consistency4.0
  • Camera & framing2.5
  • Quality & detail4.5
  • Look & lighting4.5
  • Realism & physics4.0
  • Text rendering—
  • Audio—

Polished progression, but spoon tasting is unclear and speech is absent.

Happy Horse 1.0
3.6videomodels.ai score
  • Prompt adherence3.0
  • Motion & consistency4.0
  • Camera & framing2.0
  • Quality & detail4.5
  • Look & lighting4.5
  • Realism & physics4.0
  • Text rendering—
  • Audio—

Coherent chef action, but no audible requested speech.

Helios 14B
3.1videomodels.ai score
  • Prompt adherence2.5
  • Motion & consistency3.5
  • Camera & framing2.0
  • Quality & detail3.5
  • Look & lighting4.0
  • Realism & physics3.5
  • Text rendering—
  • Audio—

The taste-to-camera sequence is out of order and speech is absent.

Seedance 1.0 Pro Fast
3.7videomodels.ai score
  • Prompt adherence4.0
  • Motion & consistency4.0
  • Camera & framing3.0
  • Quality & detail4.0
  • Look & lighting3.5
  • Realism & physics3.5
  • Text rendering—
  • Audio—

The full action sequence is clear, but lighting is less warm and details look synthetic.

Wan 2.6
3.5videomodels.ai score
  • Prompt adherence3.0
  • Motion & consistency3.5
  • Camera & framing4.0
  • Quality & detail4.0
  • Look & lighting4.5
  • Realism & physics4.0
  • Text rendering—
  • Audio2.0

Strong warm kitchen performance, but the clip has no speech track.

Kling 3.0 Omni
3.9videomodels.ai score
  • Prompt adherence4.0
  • Motion & consistency4.0
  • Camera & framing3.5
  • Quality & detail4.0
  • Look & lighting4.0
  • Realism & physics4.0
  • Text rendering—
  • Audio—

Clear taste-to-camera progression with minor mouth continuity softness.

P-Video
4.1videomodels.ai score
  • Prompt adherence4.5
  • Motion & consistency4.0
  • Camera & framing4.0
  • Quality & detail4.0
  • Look & lighting4.0
  • Realism & physics4.0
  • Text rendering—
  • Audio—

The full taste, turn, and apparent speech sequence is clearly staged.

Kling 2.5 Turbo
3.6videomodels.ai score
  • Prompt adherence3.0
  • Motion & consistency4.0
  • Camera & framing2.0
  • Quality & detail4.5
  • Look & lighting4.5
  • Realism & physics4.0
  • Text rendering—
  • Audio—

Convincing tasting shot, but no spoken line is present.

Seedance 1.0 Pro
4.0videomodels.ai score
  • Prompt adherence4.5
  • Motion & consistency4.0
  • Camera & framing3.5
  • Quality & detail4.0
  • Look & lighting4.0
  • Realism & physics3.5
  • Text rendering—
  • Audio—

Clear tasting-to-camera sequence with minor utensil and mouth continuity artifacts.

PixVerse V4.5
4.3videomodels.ai score
  • Prompt adherence4.5
  • Motion & consistency4.5
  • Camera & framing4.0
  • Quality & detail4.0
  • Look & lighting4.5
  • Realism & physics4.0
  • Text rendering—
  • Audio—

Smooth, readable taste-to-address progression with a strong warm kitchen presentation.

PixVerse V4
4.1videomodels.ai score
  • Prompt adherence4.5
  • Motion & consistency4.0
  • Camera & framing3.5
  • Quality & detail4.5
  • Look & lighting4.0
  • Realism & physics4.0
  • Text rendering—
  • Audio—

The tasting-to-speaking chain is especially clear, despite an unrequested subtitle.

Hailuo 02
3.6videomodels.ai score
  • Prompt adherence3.0
  • Motion & consistency3.5
  • Camera & framing4.0
  • Quality & detail4.0
  • Look & lighting4.0
  • Realism & physics3.5
  • Text rendering—
  • Audio—

The chef completes the visible taste-and-address chain, but it lacks the required spoken track.

Hailuo 01 Director
3.5videomodels.ai score
  • Prompt adherence3.0
  • Motion & consistency3.5
  • Camera & framing4.0
  • Quality & detail3.5
  • Look & lighting4.0
  • Realism & physics3.5
  • Text rendering—
  • Audio—

The tasting-to-camera sequence is clear, but the required spoken audio is absent.