UniAll AI SEO/GEO · Veo3.1 Video · 2026-05-31

What is Veo3.1 Video?

Veo3.1 Video is an AI video generation model available through UniAll AI. It supports multiple short-form video workflows, including text-to-video, image-to-video, first/last-frame generation, multi-reference-image generation, and video-to-video continuation or editing.

Public model id: `veo3.1-video-720p`

The model is designed for async video generation, where you submit a job, receive a task response, and then poll or retrieve the final result when processing is complete.

Who should use the Veo3.1 Video API?

The Veo3.1 Video API is useful for:

• Developers building AI video tools or creative automation workflows
• Marketing teams generating short ad concepts or product clips
• E-commerce teams creating product showcase videos
• Social media teams producing 9:16 vertical video assets
• Agencies and SaaS platforms that need programmatic video generation
• Internal content teams testing storyboards, camera motion, and visual concepts

It is best suited for short clips from 4 to 8 seconds, with one generated video per request.

Supported generation modes

UniAll AI exposes Veo3.1 Video through `/v1/videos/generations` with the `generation_mode` parameter.

| Mode | Use case | Key inputs | |---|---|---| | `text_to_video` | Generate a video from a prompt | `prompt` | | `reference_image_to_video` | Animate a single reference image | `prompt`, `image_url` | | `reference_media_to_video` | Generate from reference media context | `prompt`, optional media fields | | `first_last_frame` | Control the first and ending frame | `prompt`, `first_image_url`, `last_image_url` | | `video_to_video` | Edit, continue, or transform a source video | `prompt`, `video_url` |

Core parameters

Common request fields include:

• `model`: Use `veo3.1-video-720p`
• `generation_mode`: Select the workflow mode
• `prompt`: Required text instruction
• `duration`: 4, 6, or 8 seconds
• `aspect_ratio`: `16:9` or `9:16`
• `resolution`: `720p`, `1080p`, or `4k`, depending on selected variant availability
• `generate_audio`: Boolean toggle for audio generation when supported
• `image_url`: Single reference image
• `reference_image_urls`: 2–3 reference images
• `first_image_url` and `last_image_url`: Start and end frames
• `video_url`: Source video for video-to-video workflows

Supported upload formats include PNG, JPEG, and WebP for images, and MP4, QuickTime, or WebM for video.

Example: text-to-video request

```bash curl -X POST "https://api.uniall.ai/v1/videos/generations" \ -H "Authorization: Bearer YOUR_API_KEY" \ -H "Content-Type: application/json" \ -d '{ "model": "veo3.1-video-720p", "generation_mode": "text_to_video", "prompt": "A cinematic product reveal of a wireless headphone on a glossy black table, soft studio lighting, slow dolly-in camera movement.", "duration": 8, "aspect_ratio": "16:9", "resolution": "720p", "generate_audio": true }' ```

Example: image-to-video request

```bash curl -X POST "https://api.uniall.ai/v1/videos/generations" \ -H "Authorization: Bearer YOUR_API_KEY" \ -H "Content-Type: application/json" \ -d '{ "model": "veo3.1-video-720p", "generation_mode": "reference_image_to_video", "prompt": "Animate the product with subtle camera orbit, premium lighting, and realistic reflections. Keep the object shape consistent.", "image_url": "https://example.com/product.png", "duration": 6, "aspect_ratio": "9:16", "resolution": "720p" }' ```

Example: first and last frame video

```json { "model": "veo3.1-video-720p", "generation_mode": "first_last_frame", "prompt": "Create a smooth transition from morning city skyline to sunset skyline, cinematic and realistic.", "first_image_url": "https://example.com/start.png", "last_image_url": "https://example.com/end.png", "duration": 8, "aspect_ratio": "16:9", "resolution": "720p" } ```

Pricing and billing angle

Veo3.1 Video is billed by generated video seconds. For the `veo3.1-video-720p` standard 720p variant, the listed user price is USD $0.136 per second. Audio-enabled and higher-resolution variants may use different per-second rates.

Because billing is time-based, your cost is mainly affected by:

• Clip duration: 4s, 6s, or 8s
• Resolution: 720p, 1080p, or 4K
• Whether audio generation is enabled
• Standard vs fast model variants, when available

For predictable spend, start with 4-second or 6-second drafts, then regenerate final candidates at the desired duration and resolution.

Prompting tips for better results

Use prompts that describe the subject, motion, camera, lighting, style, and constraints. For example:

```text A realistic 9:16 vertical video of a luxury skincare bottle on wet marble, slow push-in camera, soft diffused light, water droplets, premium commercial style, no text overlays. ```

Good prompts often include:

• Subject: what should appear
• Scene: where it happens
• Motion: camera and subject movement
• Style: cinematic, commercial, documentary, anime, etc.
• Constraints: keep logo visible, avoid extra text, preserve product shape

Implementation notes

• The API is asynchronous, so design your app around job submission and result retrieval.
• Use public HTTPS URLs for image and video inputs.
• Validate duration, aspect ratio, and resolution before submitting jobs.
• For reference-image workflows, provide 2–3 images when using `reference_image_urls`.
• For production systems, store request ids, prompt metadata, cost estimates, and output URLs for auditing.

When to choose Veo3.1 Video

Choose Veo3.1 Video when you need high-quality short clips, product visuals, ad concepts, vertical social content, or controlled generation using reference media. If latency or cost is the top priority, compare available fast variants and lower-resolution settings before scaling.