A wide, locked studio shot shows three dancers performing eight synchronized counts: step forward, heel turn, low squat, arm sweep, then a fast half-spin and precise final pose. Loose jackets and long skirts follow the motion naturally. Polished floor reflections, evenly timed beat, full bodies and feet always visible.

Music & Dance · default settings, one run, 5 s, 720p, 16:9 · Updated September 15, 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
3.6videomodels.ai score
  • Prompt adherence2.0
  • Motion & consistency4.0
  • Camera & framing4.5
  • Quality & detail4.0
  • Look & lighting4.0
  • Realism & physics4.0
  • Text rendering—
  • Audio—

Four dancers and missing heel-turn and low-squat steps break the requested choreography.

Seedance 2.5
3.6videomodels.ai score
  • Prompt adherence3.0
  • Motion & consistency3.0
  • Camera & framing4.5
  • Quality & detail4.0
  • Look & lighting4.5
  • Realism & physics3.5
  • Text rendering—
  • Audio—

The spin and reflections are strong, but the sequence order and jacket requirement miss.

Seedance 2.0
4.3videomodels.ai score
  • Prompt adherence4.5
  • Motion & consistency4.0
  • Camera & framing5.0
  • Quality & detail4.0
  • Look & lighting4.5
  • Realism & physics4.0
  • Text rendering—
  • Audio—

The full synchronized sequence is clearly visible, with only slight motion softness during the spin.

Seedance 2.0 Fast
3.5videomodels.ai score
  • Prompt adherence3.0
  • Motion & consistency3.0
  • Camera & framing4.5
  • Quality & detail4.0
  • Look & lighting4.5
  • Realism & physics3.0
  • Text rendering—
  • Audio—

The squat, sweep and turn read, but the heel turn and final pose are indistinct.

Kling 3.0 Turbo
3.7videomodels.ai score
  • Prompt adherence2.5
  • Motion & consistency3.5
  • Camera & framing4.5
  • Quality & detail4.5
  • Look & lighting4.0
  • Realism & physics4.0
  • Text rendering—
  • Audio4.0

Strong locked studio framing, but the heel turn and fast half-spin are not clear.

Veo 3.1
3.6videomodels.ai score
  • Prompt adherence2.5
  • Motion & consistency3.0
  • Camera & framing4.5
  • Quality & detail4.5
  • Look & lighting4.5
  • Realism & physics4.0
  • Text rendering—
  • Audio4.0

The squat is clear, but the required heel turn and half-spin never read.

Veo 3.1 Fast
3.3videomodels.ai score
  • Prompt adherence2.5
  • Motion & consistency2.5
  • Camera & framing4.0
  • Quality & detail4.0
  • Look & lighting4.0
  • Realism & physics4.0
  • Text rendering—
  • Audio4.0

Clean framing and a squat, but the phrase stalls without the requested heel turn or half-spin.

Happy Horse 1.1
4.0videomodels.ai score
  • Prompt adherence4.0
  • Motion & consistency3.5
  • Camera & framing4.5
  • Quality & detail4.5
  • Look & lighting4.0
  • Realism & physics3.5
  • Text rendering—
  • Audio4.0

This is the clearest full choreography, though the heel turn is not specifically legible.

MiniMax H3
3.7videomodels.ai score
  • Prompt adherence3.5
  • Motion & consistency3.5
  • Camera & framing4.0
  • Quality & detail4.0
  • Look & lighting3.5
  • Realism & physics4.0
  • Text rendering—
  • Audio4.0

The dance sequence is well covered, but one dancer lacks a long skirt and reflections are weak.

Gemini Omni Flash
4.1videomodels.ai score
  • Prompt adherence4.0
  • Motion & consistency4.0
  • Camera & framing4.5
  • Quality & detail4.0
  • Look & lighting4.5
  • Realism & physics4.0
  • Text rendering—
  • Audio4.0

Clean locked-off trio sequence, though the exact heel turn is unclear.

Wan 2.7
3.7videomodels.ai score
  • Prompt adherence3.5
  • Motion & consistency4.0
  • Camera & framing4.5
  • Quality & detail4.0
  • Look & lighting4.5
  • Realism & physics4.0
  • Text rendering—
  • Audio1.0

Heel turn and half-spin are not clearly readable.

Grok Imagine Video
3.8videomodels.ai score
  • Prompt adherence3.5
  • Motion & consistency3.5
  • Camera & framing4.5
  • Quality & detail4.0
  • Look & lighting4.0
  • Realism & physics4.0
  • Text rendering—
  • Audio4.0

The sequence and final pose read well, but the garments are robes rather than jackets and skirts.

P-Video 2
3.9videomodels.ai score
  • Prompt adherence4.0
  • Motion & consistency3.5
  • Camera & framing4.5
  • Quality & detail4.0
  • Look & lighting4.0
  • Realism & physics3.5
  • Text rendering—
  • Audio4.0

Choreography is clear, but spin frames soften and warp slightly.

Runway Gen-4.5
3.5videomodels.ai score
  • Prompt adherence2.5
  • Motion & consistency3.5
  • Camera & framing4.5
  • Quality & detail4.0
  • Look & lighting4.0
  • Realism & physics3.5
  • Text rendering—
  • Audio—

It uses five dancers and does not clearly follow the requested footwork chain.

Hailuo 2.3
3.6videomodels.ai score
  • Prompt adherence3.0
  • Motion & consistency4.0
  • Camera & framing4.0
  • Quality & detail3.5
  • Look & lighting3.5
  • Realism & physics3.5
  • Text rendering—
  • Audio—

The trio dances coherently, but the requested step order is only partly evidenced.

Kling 2.6
3.8videomodels.ai score
  • Prompt adherence3.0
  • Motion & consistency4.0
  • Camera & framing4.5
  • Quality & detail4.0
  • Look & lighting4.0
  • Realism & physics4.0
  • Text rendering—
  • Audio—

Stable, realistic trio motion, but several specified footwork beats are not clear.

FLUX 3
4.1videomodels.ai score
  • Prompt adherence4.0
  • Motion & consistency4.0
  • Camera & framing4.5
  • Quality & detail4.0
  • Look & lighting4.0
  • Realism & physics4.0
  • Text rendering—
  • Audio4.0

A clean synchronized trio sequence with the exact heel turn not clearly visible.

PixVerse V5
4.0videomodels.ai score
  • Prompt adherence3.5
  • Motion & consistency4.0
  • Camera & framing4.5
  • Quality & detail4.0
  • Look & lighting4.5
  • Realism & physics4.0
  • Text rendering—
  • Audio—

The heel turn, low squat, and half-spin are only partly legible.

Seedance 1.5 Pro
4.1videomodels.ai score
  • Prompt adherence3.5
  • Motion & consistency4.5
  • Camera & framing4.5
  • Quality & detail4.0
  • Look & lighting4.0
  • Realism & physics4.0
  • Text rendering—
  • Audio4.0

Strong locked-off, full-body dance coverage but the exact step chain is incomplete.

FastH3
3.8videomodels.ai score
  • Prompt adherence3.0
  • Motion & consistency4.0
  • Camera & framing4.5
  • Quality & detail4.0
  • Look & lighting4.0
  • Realism & physics4.0
  • Text rendering—
  • Audio4.0

Strong synchronized trio and full chain, but the heel turn is not clearly isolated.

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

The requested low squat is not clearly shown.

Lance
3.4videomodels.ai score
  • Prompt adherence2.5
  • Motion & consistency3.5
  • Camera & framing4.5
  • Quality & detail3.5
  • Look & lighting3.5
  • Realism & physics3.5
  • Text rendering—
  • Audio—

Four dancers and a soft ending miss the requested trio and precise final pose.

Wan T2V
3.3videomodels.ai score
  • Prompt adherence2.0
  • Motion & consistency3.5
  • Camera & framing4.5
  • Quality & detail3.5
  • Look & lighting3.5
  • Realism & physics3.5
  • Text rendering—
  • Audio—

It renders five dancers instead of the requested trio.

HunyuanVideo 1.5
3.5videomodels.ai score
  • Prompt adherence2.5
  • Motion & consistency3.5
  • Camera & framing4.0
  • Quality & detail4.0
  • Look & lighting4.0
  • Realism & physics4.0
  • Text rendering—
  • Audio—

The requested low squat never occurs, leaving the choreography incomplete.

Happy Horse 1.0
4.2videomodels.ai score
  • Prompt adherence4.5
  • Motion & consistency4.0
  • Camera & framing4.0
  • Quality & detail4.0
  • Look & lighting4.5
  • Realism & physics4.5
  • Text rendering—
  • Audio4.0

The complete sequence is clear, though the final pose feels slightly mid-motion.

Helios 14B
3.4videomodels.ai score
  • Prompt adherence1.5
  • Motion & consistency4.0
  • Camera & framing4.5
  • Quality & detail4.0
  • Look & lighting4.0
  • Realism & physics4.0
  • Text rendering—
  • Audio—

Five dancers repeat a low-squat phrase instead of the requested three-dancer chain.

Seedance 1.0 Pro Fast
3.8videomodels.ai score
  • Prompt adherence3.0
  • Motion & consistency4.0
  • Camera & framing4.5
  • Quality & detail4.0
  • Look & lighting4.0
  • Realism & physics3.5
  • Text rendering—
  • Audio—

The three-dancer sequence is coherent, but the heel turn is not clearly visible.

Wan 2.6
3.9videomodels.ai score
  • Prompt adherence3.5
  • Motion & consistency4.0
  • Camera & framing4.5
  • Quality & detail4.0
  • Look & lighting4.0
  • Realism & physics4.0
  • Text rendering—
  • Audio—

The pose resolves before rather than after the final spin.

Kling 3.0 Omni
3.8videomodels.ai score
  • Prompt adherence4.0
  • Motion & consistency3.5
  • Camera & framing4.5
  • Quality & detail3.5
  • Look & lighting4.0
  • Realism & physics3.5
  • Text rendering—
  • Audio—

Heel-turn detail is obscured by blur and skirts.

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

It uses four dancers and several exact footwork details remain unclear.

Kling 2.5 Turbo
4.1videomodels.ai score
  • Prompt adherence3.5
  • Motion & consistency4.0
  • Camera & framing4.5
  • Quality & detail4.5
  • Look & lighting4.5
  • Realism & physics4.0
  • Text rendering—
  • Audio—

Strong locked studio image, but the step and heel turn are indistinct.

Seedance 1.0 Pro
3.6videomodels.ai score
  • Prompt adherence2.5
  • Motion & consistency4.0
  • Camera & framing4.5
  • Quality & detail4.0
  • Look & lighting3.5
  • Realism & physics4.0
  • Text rendering—
  • Audio—

The heel turn and fast half-spin are absent from the otherwise clean sequence.

PixVerse V4.5
3.8videomodels.ai score
  • Prompt adherence3.0
  • Motion & consistency4.0
  • Camera & framing4.5
  • Quality & detail4.0
  • Look & lighting4.0
  • Realism & physics4.0
  • Text rendering—
  • Audio—

The low squat never appears clearly.

PixVerse V4
3.4videomodels.ai score
  • Prompt adherence2.5
  • Motion & consistency3.5
  • Camera & framing4.0
  • Quality & detail4.0
  • Look & lighting4.0
  • Realism & physics3.5
  • Text rendering—
  • Audio—

The clip shows roughly seven dancers rather than the requested three.

Hailuo 02
3.5videomodels.ai score
  • Prompt adherence3.0
  • Motion & consistency3.5
  • Camera & framing4.5
  • Quality & detail3.5
  • Look & lighting3.5
  • Realism & physics3.5
  • Text rendering—
  • Audio—

The final pose is missing after the turn.

Hailuo 01 Director
3.2videomodels.ai score
  • Prompt adherence2.0
  • Motion & consistency2.5
  • Camera & framing4.5
  • Quality & detail4.5
  • Look & lighting4.5
  • Realism & physics2.5
  • Text rendering—
  • Audio—

The trio abruptly becomes four dancers.