Best image-to-video AI models, ranked by quality
Models that animate a start image into a video. Listed: models rated on the Image-to-Video Arena that you can run on Railwail.
Best image-to-video AI models, ranked by quality
Sorted by Arena score, highest first. Models without an Arena rating follow, by price.
Median Arena score here: 1,249
| # | Model | Arena score | Price | Value | Actions |
|---|---|---|---|---|---|
| 01 | Arena1,415±4 · 554,050 votes | USD 0.060/sUSD 0.30 for 5 s | 556 pts/$+167 over median | Try | |
| 02 | Google Veo 3.1 Google DeepMindText-to-Video Arena: 1,362 | Arena1,390±9 · 53,089 votes | USD 0.48/sUSD 1.92 for 4 s | 74 pts/$+142 over median | Try |
| 03 | Google Veo 3.1 Fast Google DeepMindText-to-Video Arena: 1,358 | Arena1,372±10 · 54,380 votes | USD 0.18/sUSD 0.72 for 4 s | 171 pts/$+123 over median | Try |
| 04 | Kling v3 Kuaishou (Kling) | Arena1,354±6 · 206,221 votes | USD 0.2688/sUSD 1.344 for 5 s | 78 pts/$+106 over median | Try |
| 05 | Arena1,299±7 · 126,710 votes | USD 0.084/sUSD 0.42 for 5 s | 121 pts/$+51 over median | Try | |
| 06 | Arena1,273±8 · 34,019 votes | USD 0.18/sUSD 0.90 for 5 s | 27 pts/$+25 over median | Try | |
| 07 | Arena1,262±5 · 405,786 votes | USD 0.336/video | 40 pts/$+14 over median | Try | |
| 08 | Arena1,235±8 · 29,824 votes | USD 0.336/sUSD 1.68 for 5 s | Not clearly above the median | Try | |
| 09 | Arena1,229±8 · 29,957 votes | USD 0.060/sUSD 0.30 for 5 s | Not clearly above the median | Try | |
| 10 | Arena1,224±10 · 21,785 votes | USD 0.324/video | Not clearly above the median | Try | |
| 11 | Arena1,185±8 · 33,734 votes | USD 0.0432/sUSD 0.216 for 5 s | Not clearly above the median | Try | |
| 12 | Arena1,166±16 · 10,303 votes | USD 0.60/sUSD 4.80 for 8 s | Not clearly above the median | Try | |
| 13 | Arena1,109±16 · 9,521 votes | USD 0.216/sUSD 1.08 for 5 s | Not clearly above the median | Try | |
| 14 | Runway Gen-4 Turbo Runway | Arena1,052±13 · 6,801 votes | USD 0.060/sUSD 0.30 for 5 s | Not clearly above the median | Try |
Value = Arena points above the median of the rated models in this list per US dollar of one default run; only models whose 95 % interval lies fully above the median get a value rank.
How to read this ranking
- Arena score
- A rating from blind pairwise votes on the public Arena: people compare two anonymous models on the same prompt and pick the better result. Higher is better; ± is the 95 % interval, “votes” the number of battles.
- Price
- What one run with the model’s default settings costs on Railwail, from the same pricing rules that bill your runs (1 credit = $0.01). ≈ marks GPU-time prices: a typical run, billed by the real run time.
- Value
- Value = Arena points above the median of the rated models in this list per US dollar of one default run; only models whose 95 % interval lies fully above the median get a value rank.
- Who is listed
- Models you can run on Railwail that the Image-to-Video Arena rates.
- Rank
- The position within this list, not the rank on the arena.
Source
Quality scores: Image-to-Video Arena leaderboard, published 22 sep 2026, retrieved 24 sep 2026 from the LMArena leaderboard dataset. Used under CC BY 4.0; we show the overall score of the models we can match to a Railwail model and rank them within this list.
Questions about this ranking
Which image-to-video model has the highest Arena score on Railwail?
Grok Imagine Video by xAI with an Arena score of 1,415 (Image-to-Video Arena, published 22 sep 2026). Next: Google Veo 3.1 (1,390) and Google Veo 3.1 Fast (1,372).
Which image-to-video model is the cheapest on Railwail?
Seedance Lite by ByteDance at USD 0.0432/s. Prices come from the same pricing rules that bill your runs.
Which image-to-video model gives the most quality per dollar?
Grok Imagine Video: 167 Arena points above the median of this list (1,249) at USD 0.060/s, that is 556 points per dollar.
How is value (quality per dollar) calculated?
Value = Arena points above the median of the rated models in this list per US dollar of one default run; only models whose 95 % interval lies fully above the median get a value rank.