Text to Video Model Leaderboard

Compare the top AI video generation models across Artificial Analysis, Design Arena, and Arena AI - unified rankings, source-by-source scores, speed, and API pricing in one table.

Magic Hour usage snapshot

Server-recorded generate actions from Aug 2–29, 2026.

Attempts may include retries and requests that did not become completed projects. This is a dated snapshot, not a live query or controlled benchmark.

Models ranked

41

in this table

Leading model

Dreamina Seedance 2.0 720p

100.0 consensus

Median consensus

54.8

typical model in this list

Gap to 2nd

3.0 pts

consensus points, 1st vs 2nd

All coverage levels

Showing 41 of 41 · Snapshot 2026-04-06

Methodology

Each source uses preference data to estimate skill scores. We map ranks to percentiles and average where a model appears on multiple lists. The bar in the Consensus column is green;purple, rose, amber match Artificial Analysis, Design Arena, Arena AI columns. Speed is approximate time to the first image.

Primary sources

• Artificial Analysis
• Design Arena
• Arena AI

FAQ

Answers below use the same snapshot as the table above (as of 2026-04-06, 41 models). Figures are from our export, not live pages at Artificial Analysis, Design Arena, or Arena AI—those sites may have moved on since we built this snapshot. The Consensus column is our average of percentile ranks across benchmarks where each model appears.

Which models have the newest release dates in this snapshot?

We use each model's released field from the export. Among rows with a parseable date, the newest in this snapshot include: Dreamina Seedance 2.0 720p (2026-03-01); SkyReels V4 (2026-03-01); PixVerse V6 (2026-03-01).

Which model is “best” on this page?

By default, we sort by Consensus, so Dreamina Seedance 2.0 720p leads this snapshot at 100.0 (average percentile across benchmarks where the model appears). By Elo in the Artificial Analysis column alone, Dreamina Seedance 2.0 720p is highest at 1274. “Best” still depends on price, latency, and which benchmarks you care about—use the sortable table.

What are the top models by Elo in this snapshot?

The Elo values below are the Artificial Analysis numbers in this export (2026-04-06), not necessarily what you see on Artificial Analysis today:

1. Dreamina Seedance 2.0 720p — Elo 1274
2. SkyReels V4 — Elo 1245
3. Kling 3.0 1080p (Pro) — Elo 1241
4. PixVerse V6 — Elo 1240
5. Kling 3.0 Omni 1080p (Pro) — Elo 1231

How is text-to-video different from image-to-video?

Text-to-video models generate video clips from a text prompt alone. Image-to-video models take both a reference image and a text prompt, then animate the image into a video clip. Our table is built for text-to-video leaderboards from the sources named in the header.

How are models ranked on this page?

Each upstream source runs preference tests and publishes ranks or scores. We map those to percentiles within each benchmark, then average across benchmarks where a model appears—that is the Consensus column. Per-source columns show the ranks and scores stored in our snapshot for Artificial Analysis, Design Arena, and Arena AI. To change upstream leaderboards, participate on those sites; our table updates when we refresh the export.

Which open-weights models lead in this snapshot?

We flag open-weights rows from the export name suffix “Open Weights”. By Artificial Analysis Elo in this snapshot, the highest are:

No open-weights rows with Elo in this export.

Treat naming as a signal only—confirm license terms with each provider before production use.

What are Elo and margin of error?

Elo in our table is the value from the snapshot for the Artificial Analysis column (and similar skill estimates elsewhere). Margin of error / intervals (e.g. in CI columns) come from that same export. For how Artificial Analysis computes Elo from votes, see their methodology; our numbers stay fixed until the next snapshot refresh.