7 models · updated 2026-09-02
Text to video models: what your GPU can actually run
Video is not an image repeated. A local text to video model sees one very long sequence, and the attention cost grows with the square of its length, which is why a few extra seconds of footage can double the render time.
Common picks
Output
Memory arrangement
Weight precision
Results
| Model | Denoiser | Verdict | Arrangement | Peak VRAM | Time |
|---|---|---|---|---|---|
| Loading… | |||||
Every model in this catalog
Download sizes are the full pipeline at the precision it is published in.
LTX 2.5
21.01B denoiser · 34.14B total · 63.6 GB to download
LTX-Video
1.923B denoiser · 7.104B total · 26.5 GB to download
Wan 2.1 T2V 1.3B
1.419B denoiser · 7.227B total · 26.9 GB to download
Stable Video Diffusion XT
1.525B denoiser · 2.255B total · 8.4 GB to download
Wan 2.2 TI2V 5B
5B denoiser · 11.386B total · 31.8 GB to download
Wan 2.2 T2V A14B
14.288B denoiser · 34.384B total · 117.5 GB to download
Wan 2.1 T2V 14B
14.288B denoiser · 20.096B total · 74.9 GB to download