Describe a shot. Get a finished clip. Skrrol's text-to-video runs Sora 2 and Veo 3 inside one studio with a real editor on the back end.
Text-to-video is the most direct AI video workflow — write a shot description, get a moving clip back. Skrrol AI runs the strongest text-to-video models available, OpenAI's Sora 2 (and Sora 2 Pro for higher fidelity) and Google's Veo 3 / Veo 3.1, behind one prompt box. Pick a model, write the shot, choose an aspect ratio, hit generate, and a cinematic clip lands in your library.
Good text-to-video prompts have shape: subject ("a fox"), action ("running through tall grass"), camera ("slow tracking shot from low angle"), lighting ("golden hour, warm side light"), style ("cinematic, slight film grain"), and duration ("6 seconds"). The more specific the prompt, the more reliable the result. Skrrol's prompt panel surfaces structure hints so good prompts come naturally.
Where Skrrol differs from a one-shot generator is what happens after the clip lands. The full editor opens on top of the generation — multi-track timeline, color grading, transitions, text, music, voiceover, keyframes, chroma key. You can stack multiple AI clips into a sequence, color-match them, layer in music from the music generator, add captions, and export to MP4 in any aspect ratio. The whole pipeline — generate, edit, export — happens inside one browser tab.
Text-to-video is the heaviest paid modality on Skrrol because the underlying compute is heavy. Free tier is for learning the prompt language; Standard at €7.99 covers a creator's weekly cadence; Advanced (€16.99) and Advanced Pro (€34.99) cover daily iteration and studio-scale work. Every clip is local-first — saved to your device, not a content cloud, until export.
Pick the model per shot. Sora 2 leans cinematic and stylised; Veo 3 leans naturalistic with strong motion physics. A/B both for the same prompt.
Generate directly into 16:9, 9:16, 1:1, and 21:9 — match the platform you'll publish to without post-crop.
Pick the clip length the model supports. For longer pieces, generate multiple shots and sequence them on the timeline.
Specify camera moves (push-in, dolly, handheld), lighting (golden hour, hard rim, soft window), and stylistic references inside the prompt.
Color grade, add transitions, layer text and music, ramp speed, and export — on the same canvas the clip just landed on.
Renders are stored in OPFS on your device. Export when you want them out; otherwise nothing leaves the browser.
Result: A wide opening shot with realistic atmosphere — the kind of shot a vlog or documentary would normally license from a stock site.
Result: A vertical product hero ready for Reels, Shorts, and TikTok with on-brand cinematography.
Result: A stylised animated clip useful for kids' content, music videos, and fantasy storytelling without a 2D animation budget.
Result: A piece of B-roll that cuts cleanly into a documentary edit and tolerates voiceover.
Sign in, click Generate, pick the Video tab, and select Text-to-Video.
Sora 2 for cinematic and stylised; Veo 3 for naturalistic with motion physics. Sora 2 Pro and Veo 3.1 cost more credits for higher fidelity.
Subject, action, camera move, lighting, style, duration. Be specific. Add aspect ratio inline.
Hit Generate. The clip appears in your library when ready, with the prompt saved alongside it.
Click the clip to open it in the Skrrol editor. Trim, color-grade, add captions, and pair with music or voice.
Stack multiple AI clips, add transitions, render to MP4 in your chosen aspect ratio, and download.
Skrrol AI uses VL credits across all generators — image, video, voice, and music. The same credit pool applies; heavier modalities (video) use more credits per generation than lighter ones (image, voice). Choose a plan and use credits across any generator.
Trial credits to generate a few short test clips. a few hundred credits per clip on the higher-quality models, less for fast/preview tiers. Watermarked.
8000 VL credits — roughly the volume needed for a steady weekly upload schedule when each clip costs about a few hundred credits. No watermark.
17000 VL credits — comfortable headroom for daily iteration, B-roll generation, and ad variants without rationing prompts.
35000 VL credits — production volume for studios shipping multiple finished video pieces per week and re-rolling generations to dial in shots.
Each model has its own duration cap. Skrrol surfaces the maximum at the top of the panel for the active model. For longer videos, generate multiple shots and sequence on the timeline.
Veo 3 supports synchronised audio in many cases. For full control, generate the clip silent and add music from the music generator and voiceover from the voice generator on the timeline.
Text-to-video is the heaviest modality. A single Sora 2 or Veo 3 generation costs a meaningful fraction of a paid plan's monthly credits. Standard (€7.99) is enough for a steady creator cadence; Advanced and Advanced Pro give studio-scale headroom.
Yes — generate multiple shots, sequence them on the timeline, add transitions and music. A 60-second piece is typically 5–10 generated shots edited together.
Use image-to-video with the first generated frame (or a reference image) as a starting point for subsequent shots. Text-to-video alone will drift on identity across re-rolls.
Clips live in OPFS on your device. The prompt itself is sent to the upstream model provider so the model can run, but Skrrol doesn't retain your generated assets on a content cloud — they stay local.
Every generation opens directly in the Skrrol editor. These features are particularly useful as the next step after a text-to-video ai generator run.
Skrrol AI runs every generator next to a full pro editor. Your work stays on your device. Start free.