whatmodel.sh / video / Runway Gen-3 Alpha
Runway Gen-3 Alpha
↗ share
+ contribute
Runway Gen-3 Alpha is a closed-weights video model from Runway, released June 2024.
Model information sources
$ how it compares
The video models ranked either side of Runway Gen-3 Alpha, out of 139 with a score, then more from Runway.
| # | model | score | where | ||
|---|---|---|---|---|---|
| 99 |  Wan 2.1 T2V 1.3B | 40±4 | api $0.040/sec +local | ||
| 100 |  Mochi 1 | 40±2 | api price unknown +local | ||
| 101 | Runway Gen-3 Alpha | 40±3 | comparison only | ||
| 102 |  Open-Sora Plan v1.3 (I2V) | 39±7 | local | ||
| 103 |  UniVideo | 39±3 | comparison only | ||
| more from Runway | |||||
| 26 | Runway Aleph 2.0 | 65±7 | api $0.28/sec | ||
| 36 | Runway Gen-4.5 | 58±2 | api $0.12/sec | ||
| 41 | Runway Aleph | 56±5 | comparison only | ||
$ what it's good at
text to video▇▅▄▆█▆▇▂▃▃▂▄ #67 of 94 · 82% of its score · 3 tests
Trails Pika 2.2, LTX-2.3, T2V-Turbo-v2 here, though they rank lower overall.
| test | result | vs the others in its run | counts |
|---|---|---|---|
| AIGVE-60K human perception MOS (LOVE) | 59.2% ✎ | ▄▄·█··█▄▄▄▄▄ beats 9/10 | 39% |
| AIGVE-60K human text-video correspondence MOS (LOVE) | 55.7% ✎ | ▄▄·█▄█▄··▄▄▄ beats 8/10 | 22% |
| SkyReels-Bench T2V (human rating) | 50.6% ✎ | ▄····▄···█·▄ beats 0/4 | 21% |
image to video▁▁▂▃██▄▃▃▄▂▂ #63 of 74 * · 15% of its score · 1 test
Beats Wan 2.2 A14B, LTX-2 here, though they rank higher overall.
| test | result | vs the others in its run | counts |
|---|---|---|---|
| MagicArena image-to-video Elo (Seaweed-7B report) | 1000.0 ✎ | ▄·····▄█··▄▄ beats 2/5 | 15% |
$ where to run it
No verified API or download recorded yet · add one
$ how the score was worked out · LOVE: Benchmarking and Evalua… 61%, SkyReels-V2: Infinite-length… 21%, Seaweed-7B: Cost-Effective Tr… 15%
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
| source | results | share of the score |
|---|---|---|
 LOVE: Benchmarking and Evaluating Text-to-Video Generation and Video-to-Text Interpretation (AIGVE-60K) | 2 | 61% |
| SkyReels-V2: Infinite-length Film Generative Model | 1 | 21% |
| Seaweed-7B: Cost-Effective Training of Video Generation Foundation Model | 1 | 15% |
| prior · pull toward the middle | 3% | |
| weighted average | = 39.7 |
$ 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