What is Veo3.1 Video API?
Veo3.1 Video API on UniAll AI provides access to the public model id `veo3.1-video-720p` for AI video generation. It supports multiple production workflows, including text-to-video, image-to-video, first/last frame animation, reference-image-guided video, and video continuation or editing from a source clip.
The model is designed for short-form video generation with 4, 6, or 8 second outputs, one video per generation, and asynchronous processing through the `/v1/videos/generations` endpoint.
Key capabilities
| Capability | Supported | |---|---:| | Text to video | Yes | | Image to video | Yes | | First and last frame control | Yes | | Multiple reference images | Yes, 2–3 images | | Source video input / video edit | Yes | | Video extension workflow | Yes | | Async API | Yes | | Audio generation toggle | Yes | | Aspect ratios | 16:9 and 9:16 | | Durations | 4s, 6s, 8s | | Resolutions | 720p, 1080p, 4K variants |
Who should use it?
Veo3.1 Video is a strong fit for:
- Creative tooling platforms that need AI video generation inside an app.
- Marketing teams producing product reveals, ads, social clips, and storyboards.
- E-commerce teams turning product images into short motion assets.
- Developers building automated content pipelines with an API-first workflow.
- Agencies that need both prompt-based generation and tighter visual control using reference images or first/last frames.
If speed and cost are the main priorities, compare it with Veo3.1 Fast variants. If output quality, controlled motion, and resolution options matter more, the standard Veo3.1 Video variants are usually the better starting point.
API usage overview
The API is asynchronous and uses a generation mode to define the workflow.
**Endpoint**
```http POST /v1/videos/generations ```
**Common request fields**
```json { "model": "veo3.1-video-720p", "generation_mode": "text_to_video", "prompt": "A cinematic product reveal, soft studio lighting, smooth camera movement.", "duration": 8, "aspect_ratio": "16:9", "resolution": "720p", "generate_audio": true } ```
**Common generation modes**
- `text_to_video`: generate video from a prompt.
- `reference_image_to_video`: use one image plus a prompt.
- `reference_media_to_video`: use reference media guidance.
- `first_last_frame`: animate between a starting and ending frame.
- `video_to_video`: edit or extend from a source video.
For image inputs, supported upload formats include PNG, JPEG, and WebP. For source video input, supported formats include MP4, QuickTime, and WebM.
Pricing angle
UniAll AI prices Veo3.1 Video by output second. For the standard `veo3.1-video-720p` user profile, the listed base price starts at **$0.136 per second** for 720p video without audio. Audio-enabled and 4K variants cost more, while Veo3.1 Fast variants provide lower per-second pricing for workflows that prioritize throughput.
Example pricing direction:
- Standard 720p video: lower-cost quality baseline.
- Standard 720p with audio: higher cost due to audio generation.
- Standard 4K: higher resolution for premium output.
- Fast variants: lower cost and faster-oriented workflows.
Actual charge depends on selected duration, resolution, audio setting, and account pricing context.
Veo3.1 Video vs Fast variants
| Option | Best for | Trade-off | |---|---|---| | Veo3.1 Video standard | Higher-quality creative outputs, controlled shots, production assets | Higher cost than Fast variants | | Veo3.1 Fast | Drafting, rapid iteration, high-volume testing | May be less suitable when final output quality is the top priority | | Audio variants | Social ads, explainers, richer clips | Higher per-second cost | | 4K variants | Premium campaigns and large-format output | Higher cost and heavier workflow |
Bottom line
Use Veo3.1 Video API when you need a flexible AI video generation interface that supports prompt-only creation, image-guided motion, first/last frame control, and source-video workflows. On UniAll AI, `veo3.1-video-720p` is a practical entry point for developers who want async video generation with clear parameters and selectable quality options.
常见问题
What can Veo3.1 Video API generate?
It can generate short videos from text prompts, single reference images, multiple reference images, first and last frames, or a source video. It supports 4, 6, and 8 second outputs with 16:9 or 9:16 aspect ratios.
What is the public model id for Veo3.1 Video on UniAll AI?
The public model id is `veo3.1-video-720p`. Other resolution and speed variants are available, but this id should be used when calling the 720p standard Veo3.1 Video model.
How is Veo3.1 Video priced?
It is billed per output second. The standard 720p user price starts at $0.136 per second without audio. Audio, 4K, and some variant choices change the per-second rate.