Why did Sora win the Video — 5s text-to-video benchmark?

Sora scored highest across the 5 rated dimensions (Motion coherence, Face/character consistency, Lighting realism, Camera direction, Artifact-free). Sora is still the quality leader for text-to-video at 5 seconds, but access is limited and expensive. See the per-model scorecard above for the specific strengths and weaknesses on each dimension.

Can I trust these benchmark results?

We ran the exact same prompt 3-5 times on each model with temperature 0.7 where configurable, then averaged. Scores are editorial — honest, careful, reproducible — but not an academic study with 1,000 runs. The exact prompt is published above so you can verify yourself.

How often is this benchmark updated?

Last run: 2026-09-20. We re-run when a tested model gets a major version bump (Midjourney v6 → v7, Claude 3.5 → 4, etc.) or quarterly if nothing changes. If a new model enters the category (e.g. a new image model competing with Midjourney), it gets added to the leaderboard within 30 days of launch.

Which model is the best value for money on this task?

Kling AI 1.6. For most users, the price-performance winner matters more than the overall winner — if you're running thousands of generations a month, a 10× cheaper model that scores 1-2 points lower is often the right call.

How can I reproduce this benchmark?

The exact prompt is in the "The prompt" section above. Copy it, run it on these 5 models (Sora, Runway Gen-3, Kling AI 1.6, LTX Video, Pika 2.0) with temperature 0.7, and score each output on Motion coherence, Face/character consistency, Lighting realism, Camera direction, Artifact-free. Your results will be within one point of ours on most dimensions.

Best AI model for a 5-second text-to-video

Task: Short cinematic video from a text prompt.

Video Generation5 modelsLast run: 2026-09-20

🏆 Winner

Sora

🥈 Runner-up

Runway Gen-3

💰 Price-performance

Kling AI 1.6

Sora is still the quality leader for text-to-video at 5 seconds, but access is limited and expensive. Runway Gen-3 is the practical professional choice. For creators on a budget, Kling AI 1.6 is a legitimate second-best — the free tier produces publishable results. LTX Video is free and self-hostable but you'll do visible touch-up work.

The prompt (identical across all runs)

Copy-paste prompt

A woman walking through a Tokyo street at night, neon signs reflecting on wet pavement, slight camera dolly-forward, moody cyberpunk aesthetic, cinematic 24fps, 5 seconds, no artifacts, maintain photo-realism, consistent character.

Leaderboard

Rank	Model	Total /50	Cost / run
🥇	Sora	49	15¢
🥈	Runway Gen-3	44	10¢
🥉	Kling AI 1.6	43	1.0¢
#4	Pika 2.0	39	2.0¢
#5	LTX Video	34	Free

Fast and free (self-hosted / HF Space). Noticeable face warping. OK for b-roll, not hero shots.

Methodology

• Each model ran the exact same prompt 3-5 times (temperature 0.7 where configurable).
• We scored each run across 5 dimensions on a 1-10 scale, then averaged for the final.
• Cost is per completed run, rounded to the nearest 0.1¢.
• Last run: 2026-09-20. We re-run benchmarks when a model receives a major version bump.
• These are editorial benchmarks, not an academic study. We publish the exact prompt so you can reproduce.

FAQs — Video — 5s text-to-video

A woman walking through a Tokyo street at night, neon signs reflecting on wet pavement, slight camera dolly-forward, moody cyberpunk aesthetic, cinematic 24fps, 5 seconds, no artifacts, maintain photo-realism, consistent character.

Rank

Model

Total /50

Cost / run

🥇

Sora

15¢

🥈

Runway Gen-3

10¢

🥉

Kling AI 1.6

1.0¢

Pika 2.0

2.0¢

LTX Video

Free

Per-model scorecard

🥇 Sora

49 / 50 · 15¢ per run

Motion coherence

10/10

Face/character consistency

10/10

Lighting realism

10/10

Camera direction

10/10

Artifact-free

9/10

Industry-leading motion coherence. Character stayed consistent across all 5 seconds. Minor hand artifact at 0:03.

Runway Gen-3

44 / 50 · 10¢ per run

Motion coherence

9/10

Face/character consistency

8/10

Lighting realism

9/10

Camera direction

10/10

Artifact-free

8/10

Strong camera-direction adherence. Face swam slightly in frame 72.

Kling AI 1.6

43 / 50 · 1.0¢ per run

Motion coherence

9/10

Face/character consistency

9/10

Lighting realism

8/10

Camera direction

8/10

Artifact-free

9/10

Consistently clean output. The best free-tier option for 5s videos.

Pika 2.0

39 / 50 · 2.0¢ per run

Motion coherence

8/10

Face/character consistency

7/10

Lighting realism

8/10

Camera direction

8/10

Artifact-free

8/10

Good for stylized/animated looks, less so for photorealism. Fast turnaround.

LTX Video

34 / 50 · Free per run

Motion coherence

7/10

Face/character consistency

6/10

Lighting realism

7/10

Camera direction

7/10

Artifact-free

7/10

Fast and free (self-hosted / HF Space). Noticeable face warping. OK for b-roll, not hero shots.

Methodology

• Each model ran the exact same prompt 3-5 times (temperature 0.7 where configurable).

• We scored each run across 5 dimensions on a 1-10 scale, then averaged for the final.

• Cost is per completed run, rounded to the nearest 0.1¢.

• Last run: 2026-09-20. We re-run benchmarks when a model receives a major version bump.

• These are editorial benchmarks, not an academic study. We publish the exact prompt so you can reproduce.