# AI Video

AI video generation has changed completely. What used to take a full production team can now be done in minutes with a single text prompt.

Whether you need a product demo, a social media clip, a cinematic scene, or an animated concept - i10x.ai gives you access to the best video models available, all from one interface.

{% hint style="info" %}
💡 Video generation takes longer than image generation - typically between 30 seconds and a few minutes depending on the model and settings. Be patient, the result is worth it.
{% endhint %}

### How to Generate a Video

1. Click "Video" in the left sidebar of your i10x dashboard.
2. Select your preferred video model.
3. Write a detailed prompt describing the video you want to create.
4. Adjust the settings (aspect ratio, duration, reference images).
5. Click Generate Video and wait for your clip to render.
6. Preview, download, or regenerate with adjusted settings.

### Choosing Your Video Model

Each video model has different strengths, styles, and credit costs. Here is a breakdown to help you pick the right one:

| Model                     | Best For                                         | Highlights                                                                                  | Speed       |
| ------------------------- | ------------------------------------------------ | ------------------------------------------------------------------------------------------- | ----------- |
| **OpenAI Sora 2**         | High-quality general purpose video, synced audio | OpenAI's flagship video model, strong prompt understanding                                  | ⚡ Fast      |
| **Sora 2 Pro**            | Premium cinematic quality, complex scenes        | Highest quality Sora output, more detail and realism                                        | 🐢 Thorough |
| **Google Veo 3.1**        | Cinematic video with context-aware audio         | Higher fidelity, supports first frame + last frame interpolation and reference images (R2V) | 🐢 Thorough |
| **Google Veo 3 Standard** | Balanced quality and cost                        | Great all-rounder for most video tasks                                                      | ⚡ Fast      |
| **Google Veo 3 Fast**     | Speed-first generation                           | Quickest Veo output, ideal for drafts and iteration                                         | ⚡⚡ Fastest  |
| **Google Veo 2**          | Reliable, proven video generation                | Solid quality for everyday video tasks                                                      | ⚡ Fast      |
| **Kling v1.6 Standard**   | Creative and stylized video clips                | Strong artistic control, great for social content                                           | ⚡ Fast      |
| **Kling v1.6 Pro**        | Higher quality Kling output                      | More detail and smoother motion than Standard                                               | 🐢 Thorough |
| **Kling v2.0**            | Latest Kling generation, best quality            | Most advanced Kling model, excellent motion realism                                         | 🐢 Thorough |
| **MiniMax Video-01**      | Creative storytelling and animated content       | Unique visual style, great for narrative-driven clips                                       | ⚡ Fast      |

{% hint style="info" %}
💡 Quick Guide:

→ Best overall quality? Use Sora 2 Pro or Google Veo 3.1.\
→ Need audio in your video? Use Google Veo 3.1 or OpenAI Sora 2 (synced audio).\
→ Want to animate an existing image? Use Google Veo 3.1 with a reference image.\
→ Creating social media content? Try Kling v1.6 Pro or MiniMax Video-01.\
→ Just testing or iterating? Use Google Veo 3 Fast to save credits.
{% endhint %}

### Understanding the Settings

Settings vary slightly between models but here is what you will commonly find across the video generator:

| Settings             | What it does                                                                                                                                          |
| -------------------- | ----------------------------------------------------------------------------------------------------------------------------------------------------- |
| **Prompt**           | Describe the video you want. Include subject, action, style, mood, camera movement and lighting for best results.                                     |
| **Duration**         | The length of the video clip in seconds. Longer duration = more credits.                                                                              |
| **Aspect Ratio**     | The shape of your video. Use 16:9 for landscape (YouTube, presentations), 9:16 for portrait (TikTok, Reels, Shorts), 1:1 for square (social media).   |
| **Input Image**      | Upload a starting image for the AI to animate from. Ideal formats are 16:9 (1280×720) or 9:16 (720×1280) depending on your aspect ratio.              |
| **Last Frame**       | Upload an ending image. The AI creates a smooth transition between your first and last frame. Available on select models.                             |
| **Reference Images** | Upload 1–3 images for subject-consistent generation (also called R2V — Reference to Video). The AI keeps your subject consistent throughout the clip. |
| **Input Reference**  | An optional image or video to guide the overall style and direction of the generation.                                                                |

{% hint style="info" %}
⚠️ Note: Reference Images only work with 16:9 aspect ratio and 8-second duration on some models. If Last Frame is provided alongside Reference Images, the Last Frame will be ignored.
{% endhint %}

### Writing a Great Video Prompt

Video prompts work best when you describe the scene like a film director. Think about:

→ Subject - Who or what is in the video?\
→ Action - What is happening?\
→ Camera - Is the camera moving? Zooming in? Panning?\
→ Style - Cinematic, documentary, animated, lo-fi?\
→ Mood - Dramatic, calm, energetic, mysterious?\
→ Lighting - Golden hour, studio lighting, neon lights?

| ❌ Weak Prompt      | ✅ Strong Prompt                                                                                                                                   |
| ------------------ | ------------------------------------------------------------------------------------------------------------------------------------------------- |
| "A person walking" | "A confident woman walking through a busy Tokyo street at night, neon reflections on wet pavement, cinematic slow motion, shallow depth of field" |
| "A product video"  | "A sleek black smartwatch rotating slowly on a white surface, soft studio lighting, close-up macro shot, premium product advertisement style"     |
| "A nature scene"   | "A drone shot slowly rising above a misty mountain forest at sunrise, golden hour light breaking through the trees, cinematic 4K quality"         |

### Pricing

Each video model has a different credit cost depending on the model, duration, and quality settings you choose. You can see the exact credit cost and estimated number of runs at the bottom of each model's settings panel before you hit Generate Video.

{% hint style="info" %}
💡 Credit Tip: Use Google Veo 3 Fast to test your prompt and get the motion and composition right first. Once you are happy with the direction, switch to Sora 2 Pro or Google Veo 3.1 for the final high-quality render. This saves significant credits during iteration.
{% endhint %}

{% hint style="info" %}
✅ Try This Now:

Select Google Veo 3 Fast and type:

"A slow motion close-up of coffee being poured into a white cup, steam rising, warm golden lighting, cinematic style."

Hit Generate Video and see what i10x.ai creates in minutes.
{% endhint %}


---

# Agent Instructions: Querying This Documentation

If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://knowledge.i10x.ai/ai-video.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.