Details
The Civitai video generator integrates multiple third-party text-to-video and image-to-video models directly into the same on-site interface used for image generation. Users enter a text prompt or upload an image and select a video model; the platform handles all compute and returns a short video clip. Supported models have included Lightricks LTXV, Hailuo (MiniMax), Kling (Kuaishou), Hunyuan (Tencent), Wan, and Google DeepMind's Veo 3. Civitai also introduced an on-site video LoRA trainer, allowing users to fine-tune video generation models. Generation costs Buzz.
Have evidence about Civitai's AI practices? Submit a report.
Report a Sighting →