We kept hitting the same wall in our own workflow. Making one 30-second clip meant Runway for the base video, a separate tool for lip sync, a third for upscaling, a fourth for the voiceover. Four bills, four asset libraries, four places to lose track of a project. And every time a new model dropped — Sora 2, Veo 3.1, Kling 3.0 — that was a fifth. So we started building an internal wrapper just to stop switching tabs. That turned into Vdoo AI. The part that surprised us: we assumed people wanted "the best model." They don't. They want to run four models on the same prompt and pick the one that got the face right. Once that clicked, the product changed shape — we stopped hiding the model picker, made cross-engine comparison the default, and made the asset library shared across engines instead of per-model. Where it landed: • 20+ models in one workspace — Sora 2 Pro, Veo 3.1, Kling 3.0, Seedance, Runway, Midjourney, Nano Banana 2, Hailuo 2.3, Flux • Full pipeline — text-to-video, image-to-video, motion control, lip sync, face swap, video extend, 4K super-resolution, AI audio • Credits instead of subscriptions. They don't expire, and you decide which model to spend them on Still rough in places and we know it. What I most want to hear: which model combination you actually reach for, and what's missing from the pipeline. New accounts get [N] free credits, no card needed. PH folks — code PRODUCTHUNT for [X]% off. I'll be in the comments all day.

Vdoo AI

Sora 2, Veo 3.1, Kling & 20+ AI models in one workspace

Sora 2, Veo 3.1, Kling & 20+ AI models in one workspace