whatmodel.sh / video / Dreamina Seedance 2.0
Dreamina Seedance 2.0
↗ share
+ contribute
Dreamina Seedance 2.0 is a closed-weights video model from ByteDance Seed, released March 2026. Licence: proprietary.
$ how it compares
The video models ranked either side of Dreamina Seedance 2.0, out of 139 with a score, then more from ByteDance Seed.
| # | model | score | where | ||
|---|---|---|---|---|---|
| 3 |  Gemini Omni Flash | 80±2 | api $0.10/sec | ||
| 4 |  Wan 3.0 | 79±2 | api $0.10/sec | ||
| 5 | Dreamina Seedance 2.0 | 79±1 | api $0.30/sec | ||
| 6 |  MiniMax H3 | 78±1 | api $0.13/sec +local | ||
| 7 |  Dreamina Seedance 2.5 | 77±1 | api $0.30/sec | ||
| more from ByteDance Seed | |||||
| 39 | Seedance 1.5 Pro | 58±2 | api $0.052/sec | ||
| 43 | Seedance 1.0 Pro | 56±2 | api $0.052/sec | ||
| 65 | Seaweed-7B | 50±6 | comparison only | ||
$ what it's good at
text to video▆▄▃▄▆▅█▃▂▂▂▄ #4 of 68 * · 30% of its score · 2 tests
Beats Wan 3.0 here, though they rank higher overall. Trails FLUX 3 Video here, though they rank lower overall.
| test | result | vs the others in its run | counts |
|---|---|---|---|
| LMArena video | 1479.2 ✎ | ▂▂▃▆█▆·▆▅▁▂▇ beats 38/41 | 28% |
| Sci-VBench testmini (expert human rating) | 58.8% ✎ | ▅·▃█▃·▅▃▅▃▃▃ beats 12/14 | 2.5% |
image to video▁▂·▂▅█▅▄▅▆▃▄ #4 of 50 · 56% of its score · 6 tests
Beats Wan 3.0 here, though they rank higher overall. Trails MiniMax H3 here, though they rank lower overall.
| test | result | vs the others in its run | counts |
|---|---|---|---|
| WorldReasonBench human Elo (image+text-to-video world-state prediction) | 1471.0 ✎ | ▃▃··█··▃▃▃·▃ beats 8/8 | 19% |
| Arena image-to-video | 1474.6 ✎ | ▁▁▁▁▄▇▃▆▃▂▄█ beats 36/40 | 18% |
| Rapidata physics realism (I2V, with reference) | 1268.0 ✎ | ▂··▂▂▆█▆▂▂▆▄ beats 16/19 | 11% |
| VGI-Bench visual reasoning (I2V) | 51.0% ✎ | ▄▄···▄▄█·█·▄ beats 8/8 | 3.8% |
| VWG-Bench think-with-video reasoning (overall) | 70.4% ✎ | █··▄······▄█ beats 4/5 | 2.4% |
| What-If World causal-intervention video generation (APEO paired average, Gemini 3.1 Pro judge) | 41.8% ✎ | █·▄····▄···█ beats 3/5 | 1.7% |
video editing▄··█··▄··█▄█ #5 of 9 * · 5.7% of its score · 1 test
Trails Dreamina Seedance 2.5, MiniMax H3 here, though they rank lower overall.
| test | result | vs the others in its run | counts |
|---|---|---|---|
| Arena video edit | 1364.6 ✎ | █··█··▄··█▄█ beats 5/9 | 5.7% |
other tests3 not in a task · 7.1% of its score
| test | result | vs the others in its run | counts |
|---|---|---|---|
| SWITCH interactive-interface I2V (VLM-judge overall) | 52.8% ✎ | ███··█·····█ beats 4/4 | 4.7% |
| VideoArgus-Bench TI2V (rubric-grounded score) | 89.7% ✎ | █··▄▄▄▄····▄ beats 6/6 | 1.2% |
| VideoArgus-Bench T2V (rubric-grounded score) | 93.4% ✎ | █▄·▄█▄▄····▄ beats 8/8 | 1.2% |
$ where to run it
Through an API
api · fal.ai · $0.30/sec · fal.ai ↗ · price sourceUSD per second of generated 720p video on fal.ai (standard, not fast tier). Product: fal Seedance 2.0 text-to-video endpoint. API model: bytedance/seedance-2.0/text-to-video. Settings: resolution=720p (endpoint default), audio on or off (fal states the cost is the same either way). Price basis: per-second figure printed by fal; fal bills $0.014 per 1,000 tokens underneath. 1080p $0.682/s; fast tier 720p $0.2419/s. Not the BytePlus direct tariff. Currency: USD. fal.ai model page, observed 2026-09-25.
$ how the score was worked out · LMArena leaderboard dataset 52%, WorldReasonBench: Human-Align… 19%, Rapidata/world-model-physics… 11% and 6 more
Each result implies a score on its own, after adjusting for how hard its test is and who ran it. The score is their weighted average, with a small pull toward the middle for thinly tested models. Each test's weight is its counts column above; here it's added up by who ran the tests. How it works
$ improve this page
Use a short GitHub form or edit the data through a pull request. Every number needs a source; here's which ones we can use.
data updated · sources and licences · json