whatmodel.sh / video / Wan2.1-VACE-14B
Wan2.1-VACE-14B
↗ share
+ contribute
Wan2.1-VACE-14B is an open-weights video model from Alibaba, released May 2025. Its published size is 14B parameters. Licence: apache-2.0.
Catalogue identity notes
Alibaba's VACE all-in-one video creation/editing model on Wan2.1 14B (Wan-AI/Wan2.1-VACE-14B, pinned 539c162; repo created 2025-05-13). WROP Table 1 lists 'Wan-VACE 14B', open weights, run locally on multiple GPUs as an edit/transfer (repaint) generator at 832x480, 16 fps, 57 frames. There is one 14B VACE release on Wan2.1 (the 1.3B is separate). Comparison-only: local route not verified in this pass.
Model information sources
Release date ↗ · Weights licence ↗
14B as named by Alibaba (model name, Wan2.1 14B backbone); VACE context blocks, text encoder and VAE not counted separately.
$ how it compares
The video models ranked either side of Wan2.1-VACE-14B, out of 139 with a score, then more from Alibaba.
| # | model | score | where | ||
|---|---|---|---|---|---|
| 27 |  Sora 2 | 65±2 | comparison only | ||
| 28 | Wan 2.6 | 65±1 | api $0.10/sec | ||
| 29 | Wan2.1-VACE-14B | 65±7 | comparison only | ||
| 30 |  Veo 3 | 64±2 | comparison only | ||
| 31 |  Kling O3 Pro | 64±4 | api $0.11/sec | ||
| more from Alibaba | |||||
| 2 | Wan 3.0 Prime | 80±7 | api $0.14/sec | ||
| 4 | Wan 3.0 | 79±2 | api $0.10/sec | ||
| 9 | HappyHorse 1.1 | 76±3 | api $0.14/sec | ||
$ what it's good at
other tests1 not in a task · 81% of its score
| test | result | vs the others in its run | counts |
|---|---|---|---|
| WROP object-permanence exam (human Elo) | 1506.7 ✎ | ▄····██···▄█ beats 3/7 | 81% |
$ where to run it
No verified API or download recorded yet · add one
Weights and documentation: huggingface.co/Wan-AI ↗
$ how the score was worked out · Training Object Permanence in… 81%
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 |
|---|---|---|
 Training Object Permanence in World Models (WROP exam; arXiv:2609.28654v1) | 1 | 81% |
| prior · pull toward the middle | 19% | |
| weighted average | = 64.5 |
$ 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