How to Use Veo 3.1 in 2026: Access, Prompts, and Cost

To use Veo 3.1, open a text-to-video tool, describe the shot you want, choose Veo 3.1 as the model, set your aspect ratio and duration, and generate — the model returns a short, cinematic clip, and unlike most video engines it can also produce synced native audio (dialogue, sound effects, ambience) in the same pass. You do not need a separate Google subscription if you reach Veo 3.1 through a multi-model platform: on Wandura's text-to-video tool it sits alongside other video models on one credit balance, so you pay per generation instead of a monthly per-model fee.

This guide covers the fastest way to access Veo 3.1, how to prompt it well, and exactly what it costs.

What Veo 3.1 is good at

Veo 3.1 is part of Google's Veo model family, built for high-fidelity text-to-video with strong prompt adherence, realistic motion, and — its signature strength — native audio generated together with the picture, so a single prompt can return a clip that already has dialogue, effects, and ambient sound in sync. It shines on dialogue-driven scenes, cinematic establishing shots, and product moments where believable sound sells the shot as much as the motion. Because it lives next to other engines in one AI video generator, you can start on Veo 3.1 and A/B the same prompt on Kling 3.0 without switching apps.

How to use Veo 3.1, step by step

1. Open a text-to-video tool. On Wandura, go to text-to-video — no Google account required.
2. Write a clear prompt. Name the subject, the action, the camera move, and the style in one line — and describe the sound you want, since Veo 3.1 can generate audio (see prompting tips below).
3. Select Veo 3.1 as the model. Switch the model picker to Veo 3.1; the same prompt box works across models so you can compare later.
4. Set resolution and duration. Choose 720p, 1080p, or 4K and 4, 6, or 8 seconds; keep the first clip short and at 720p to iterate cheaply.
5. Generate and review. The clip is produced asynchronously; when it's ready, play it back with sound, then download it or tweak the prompt and re-run.
6. A/B against another model (optional). Re-run the same prompt on Kling from the same balance and keep whichever nails the shot.

Prompting tips for Veo 3.1

Veo 3.1 rewards specific, visual prompts — and because it generates audio, it rewards specific sound cues too. A useful pattern is subject + action + camera + lighting/style + audio: "a barista pulls an espresso shot on a marble counter, slow push-in, warm morning light, shallow depth of field, ambient café chatter and the hiss of the steam wand" beats "a coffee shop". Describe motion explicitly ("slow dolly in", "handheld pan") and keep one main action per clip — stacking several actions muddies the result. If you want spoken lines, put the dialogue in quotes and name the speaker so the model syncs it. Generate two or three variations before committing.

How much does Veo 3.1 cost?

On Wandura, Veo 3.1 is credit-based (1 credit = $0.01) and billed per generation — the exact cost is shown on the Generate button before you confirm, so nothing is hidden. Veo 3.1 produces clips of 4, 6, or 8 seconds at 720p, 1080p, or 4K. Here's what a clip actually costs:

Prices come straight from our live price list — the exact cost is always shown on the Generate button before you confirm; see pricing for current packs and plans.

Because you pay in credits per generation rather than a fixed monthly per-model fee, you only spend on the clips you actually render — draft cheaply at 720p, then re-render the winner at 1080p or 4K. All the top video models draw from the same balance, so you can A/B by model on a per-shot basis.

Veo 3.1 vs other video models

Veo 3.1 is one of several frontier models, and its native audio makes it the natural pick when sound is part of the shot. If you're deciding between engines, our AI video model comparison breaks down motion, realism, audio, and cost — and the Kling 3.0 model page shows what it can do inside Wandura. Our guide on how to use Veo 3 explains how Veo 3 relates to Veo 3.1, and the Veo 3.1 model page runs the model in its composer.