Video generators on CUDA
VRAM thresholds, backends and licenses come from the tool fact sheets: every value is collected by code from an official source and stamped with the date it was checked. Requirements marked with an asterisk * are the authors’ own description, not our measurement. Details are on each tool’s page.
Showing 7 of 73
| Tool | Type | Runs | VRAM | Backends | Formats |
|---|---|---|---|---|---|
| Allegrotext-to-video | Video | open source | from 9.3 GB (9.3–27.5 GB) * | CUDA | — |
| CogVideoXtext-to-video | Video | open source | not stated | CUDA | — |
| FramePack | Video | open source | from 6 GB * | CUDA | — |
| HelixWorld | Video | open source | from 80 GB * | CUDA | — |
| HunyuanVideotext-to-video | Video | open source | from 45 GB (45–60 GB) * | CUDA | — |
| Latte | Video | open source | not stated | CUDA | — |
| SkyReels | Video | open source | from 18.5 GB * | CUDA | — |
Cloud services — no VRAM needed
They compute on their own servers, so a hardware filter does not apply to them.
| Dream MachineMultimodal generation and editing of images, video, audio, and text | Video | cloud | cloud | not needed | MP4 |
| Hailuotext-to-speech, voice cloning, text-to-video, text-to-image, voice design | Video | cloud | cloud | not needed | — |
| KlingVideo and image generation | Video | cloud | cloud | not needed | — |
| PikaVideo generation: text-to-video, image-to-video, video-to-video | Video | cloud | cloud | not needed | — |
| RunwayVideo, image, speech, sound effects generation, voiceover, voice isolation, upscaling, avatars | Video | cloud | cloud | not needed | MP4, MOV, ZIP, WAV, PNG, EXR |
| Soratext-to-video, image-to-video, video-to-video | Video | cloud | cloud | not needed | MP4 |
| Veotext-to-video | Video | cloud | cloud | not needed | — |
| ViduVideo generation: text-to-video, image-to-video, reference-to-video, start-end video, upscaling | Video | cloud | cloud | not needed | png, jpeg, jpg, webp |
Requirements not stated by the authors
The official sources for these tools have no data for the selected filter — that means “unknown”, not “won’t work”.
| AnimateDiff | Video | open source | from 13 GB * | — | — |
| LingBot-Video | Video | open source | not stated | — | — |
| Mochi 1text-to-video | Video | open source | from 80 GB * | — | — |
| Open-Sora | Video | open source | not stated | — | — |
| Open-Sora-Plan | Video | open source | from 24 GB * | — | — |
| Stable Video Diffusionimage-to-video | Video | open source | not stated | — | — |
| VideoCrafter | Video | open source | not stated | — | — |
| ViMax | Video | open source | not stated | — | — |
| Wantext-to-video | Video | open source | from 8.19 GB * | — | — |
Fact sheets are re-checked monthly; every value on a tool page has a source link and a check date. The asterisk * means “as described by the author”: a developer’s statement, not an independent measurement.