> For the complete documentation index, see [llms.txt](https://wiki.secrets.ai/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://wiki.secrets.ai/create/video-generation.md).

# Video Generation

Choose a video model, animate a photo, control start and end frames, and guide your video with image and motion references.

{% hint style="info" %}
Video generation requires **Premium or Ultimate**. The model menu and **Generate** button show the price for your selected settings before you generate.
{% endhint %}

Open **Generate → Video**, or open a generated image and choose **Create Video**. In Studio, images and videos share the same workspace and gallery. For the full layout and gallery actions, see [Generation Studio](/create/generation-studio.md).

<figure><img src="https://3286776735-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F9ACKOAU4nFWSXPeE2B4s%2Fuploads%2Fgit-blob-1fc633dbfe4969dfd5195d94e6be95287ab34a45%2Fvideo-form.webp?alt=media" alt="Video form with Frames selected, a source image, motion prompt, Cinematic model, 720p resolution, five-second duration, and Generate price." width="480"><figcaption><p>Choose a source photo, describe the motion, and review the model, settings, and total.</p></figcaption></figure>

### Make Your First Video

1. Select a **Character** and a source photo, or use **Create Video** on an existing image to bring that photo into the form.
2. Keep **Frames** selected for a simple animation of that photo. Your source is the **Start** frame.
3. Write a **Prompt** describing what moves, how the camera moves, and any dialogue or background sound.
4. Open **Model** and choose the option that fits your scene and budget.
5. Choose **Resolution** and **Duration**. On Classic, also choose whether to include audio.
6. Check the final Moments total on **Generate**, then submit.
7. Open the result in **All** or **Videos** to play it, download it, or reuse its setup.

A simple first prompt:

> She lifts her coffee and smiles toward the window. Camera slowly moves closer. Soft cafe ambience.

### Which Model Should I Use?

| What you want                                     | Choose                    | Why                                                                       |
| ------------------------------------------------- | ------------------------- | ------------------------------------------------------------------------- |
| The highest quality and detailed prompt following | **Cinematic**             | Supports up to 4K, sound, end frames, and image/video references          |
| Faster Cinematic generation                       | **Cinematic → Fast**      | A faster option, with resolutions up to 720p                              |
| Good quality at a lower cost                      | **Everyday**              | About half the price of Cinematic at comparable settings                  |
| Intimate scenes                                   | **Intimate**              | Designed for intimate motion, with sound and clips up to 30 seconds       |
| Faster Intimate generation                        | **Intimate → Turbo**      | A speed option with an additional cost shown in the menu                  |
| The lowest-cost option                            | **Basic**                 | Sound, adjustable length and resolution, end frames, and image references |
| Simple animation with optional audio              | **More models → Classic** | Starts from one photo; turning audio off lowers the price                 |

Cinematic is a good choice for posing, undressing, and nudity. For intimate motion, choose Intimate. Pick the model for the scene you want, then compare the displayed price.

### Pick a Model

Open the **Model** control below the prompt. **Intimate**, **Cinematic**, **Everyday**, and **Basic** appear in the main list. Expand **More models** for **Classic**. Each row explains its purpose and supported options and shows a price for your settings.

<figure><img src="https://3286776735-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F9ACKOAU4nFWSXPeE2B4s%2Fuploads%2Fgit-blob-9802176db0e429c31b1a46e059661bbbeed15714%2Fvideo-menu.webp?alt=media" alt="Video model menu with Intimate and Turbo, Cinematic and Fast, Everyday, Basic, prices, and More models." width="720"><figcaption><p>Compare the model rows; use Fast or Turbo on the corresponding row.</p></figcaption></figure>

* **Fast** on Cinematic selects **Cinematic Fast**. Turn it off to return to regular Cinematic.
* **Turbo** on Intimate selects its faster variant. Check the additional cost before choosing it.
* **Classic** is under **More models**.

The new Studio puts the models in one menu; there is no separate **NSFW video** switch to reveal them. When you change models or speed options, unsupported settings may adjust automatically. Recheck the resolution, duration, reference media, and price before generating.

### All Models at a Glance

| Model                           | Resolutions           | Length | Audio    | End frame | Reference photos | Reference clips |
| ------------------------------- | --------------------- | ------ | -------- | --------- | ---------------- | --------------- |
| **Cinematic**                   | 480p, 720p, 1080p, 4K | 4–15s  | Included | Yes       | Yes              | Yes             |
| **Cinematic Fast**              | 480p, 720p            | 4–15s  | Included | Yes       | Yes              | Yes             |
| **Everyday**                    | 480p, 720p            | 4–15s  | Included | Yes       | Yes              | Yes             |
| **Intimate / Intimate + Turbo** | 480p, 720p, 1080p, 4K | 4–30s  | Included | Yes       | Yes              | Yes             |
| **Basic**                       | 480p, 720p            | 4–15s  | Included | Yes       | Yes              | No              |
| **Classic**                     | 480p, 720p, 1080p     | 4–12s  | Optional | No        | No               | No              |

Classic uses your source photo as its starting image; it does not offer the extra Frames / References options. Basic now has adjustable settings and sound, so older instructions describing it as a fixed eight-second silent clip no longer apply.

#### Example Prices

These examples use **720p, 5 seconds**, with audio on:

| Model                | Moments |
| -------------------- | ------- |
| **Cinematic**        | 1,200   |
| **Cinematic Fast**   | 1,200   |
| **Everyday**         | 640     |
| **Intimate**         | 850     |
| **Intimate + Turbo** | 1,180   |
| **Basic**            | 320     |
| **Classic**          | 600     |

Prices vary with settings. Classic costs half as much with audio off. Adding reference photos or clips does not add a separate reference charge. The price shown on **Generate** is the total to review for your actual request.

{% hint style="info" %}
**Older model names:** Advanced (Quality) is **Cinematic**; Advanced (Fast) is **Cinematic Fast**, selected using Fast on the Cinematic row. Standard is **Classic**, and Adult is **Basic**. Basic's capabilities have also expanded; use the current settings and price shown in the form.
{% endhint %}

### Make a video from a chat photo

Open a photo from your current conversation or its **Shared media** and select **Create Video**. The compact video window uses the same model menu and **Frames / References** workflow as Studio; on a phone, it opens full screen.

1. Keep the source photo and describe the action, camera movement, and sound you want. A form opened from a chat message keeps that message's photo as the start image.
2. Choose a model, then set the supported resolution, duration, end frame, or extra references. **Fast** is on the Cinematic row, **Turbo** is on Intimate, and **Classic** is under More models.
3. Review the Moments total and select **Generate**.
4. Continue chatting while the placeholder renders. The finished video appears as a companion message and in your gallery.

Select the **play button** to watch in chat, or select elsewhere on the clip to open its preview. Video Preview includes available actions such as Download, Favorite, and Publish. In a group conversation, the clip is attributed to the companion who sent its source photo. Reloading while it renders can restore the pending clip.

A video made independently in Studio stays in the gallery. **Shared media lists photos**; use the chat itself or Studio's **Videos / All** gallery to find chat videos. See [Chat Photos](/create/chat-photos.md#make-a-video-without-leaving-chat) for a screenshot and the full photo-to-video flow.

Deleting a video also removes its linked chat message. Deleting the source image removes videos made from it, including their linked messages. Download anything you want to keep before deleting.

### Reference Media & Frame Control

Use **Frames / References** to choose how the source media guides the result. These controls are available on Cinematic, Cinematic Fast, Everyday, Intimate, and Basic. Basic accepts image references, but not reference clips.

#### Frames

* **Start:** the photo the video begins from. In Studio, select it to choose or replace the source. A video form opened from a chat message keeps that message's photo fixed.
* **End (optional):** a photo to guide how the video finishes. Use it for a reveal, a change of pose, or a camera move with a clear destination.

For a loop, you can choose the same photo for Start and End. This gives the model a matching target, but does not guarantee a seamless loop. Keep the requested motion simple.

#### References

Use references when you want to borrow appearance, composition, style, or motion rather than specify the exact first and last frame.

* Up to **5 photos total**, including your source image.
* On models that support clips, up to **3 reference videos**, with a combined length of **15 seconds**.
* Basic supports the photos only. Classic supports neither extra photos nor clips.

<figure><img src="https://3286776735-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F9ACKOAU4nFWSXPeE2B4s%2Fuploads%2Fgit-blob-09b42fbfe6bfecb31b5aaceacda39a6a099d1c98%2Freferences.webp?alt=media" alt="Video References mode with an image and clip attached, a prompt using Image 1 and Video 1 mentions, and resolution, duration, and total cost controls." width="480"><figcaption><p>Attach the media first, then use its mention to say what the model should borrow.</p></figcaption></figure>

Select the **+** in the prompt's reference bar to open the picker. Use **All / Images / Videos** where available, and filter by companion, tags, recent items, or favorites. Basic's picker offers images only. On desktop, you can also drag supported gallery media onto the prompt. Remove unwanted references before submitting.

On desktop, hover over an attached reference to inspect a larger preview; video references play muted in that preview. Drag the prompt box's bottom-right corner to make more writing room.

Type **@** to choose a reference mention, such as **Image 1** or **Video 1**, or select an attached item's mention control. Match the name in your prompt to the intended item:

> @Image 1 turns toward the window, following the gentle camera movement in @Video 1. Soft cafe ambience.

Explain what each reference contributes. For example, ask for the outfit from one photo and the camera movement from a clip. Adding a reference alone does not tell the model which details matter to you.

### Pro Tip: Best Results for Intimate Scenes

You can experiment with a clip made in Intimate or Basic as a motion reference for Cinematic:

1. Make the motion with Intimate or Basic. Keep the clip within the **15-second combined reference-video limit**.
2. Choose **Cinematic**, switch to **References**, and attach the clip alongside your source photo.
3. Use the reference mentions to explain which appearance and motion to follow.

This gives Cinematic a motion example to work from; results can vary. Each new generation has its own displayed cost.

### Video Prompting Guide

{% hint style="info" %}
Prompts can be up to **5,000 characters** on Cinematic, Cinematic Fast, Everyday, and Intimate; **4,000** on Basic; and **2,000** on Classic. Cinematic has excellent prompt adherence, so the more detail you give about the motion you want, the better your results.
{% endhint %}

Cinematic is our highest-quality video model. It handles complex motion, realistic physics, multi-subject interactions, and synchronized audio generation all in a single pass. To get the most out of it, follow these prompting principles.

#### Core Prompt Structure

Since the image already provides the visual context, build your prompt around motion and change:

1. **Action & motion** — What moves and how (the most important element)
2. **Camera movement** — How the "camera" moves (one movement per shot)
3. **Pacing** — How fast or slow things happen
4. **Audio cues** — Dialogue, sounds, or atmosphere you want to hear
5. **Constraints** — What to avoid (no distortion, no jitter, etc.)

{% hint style="info" %}
**Aim for 50–200 words.** This is the sweet spot for quality. You don't need to re-describe what's already in the image — focus on the animation.
{% endhint %}

#### Key Principles

**1. Focus on motion, not the scene**

The image is your scene. Don't re-describe the setting, outfit, or appearance — the model already sees all of that. Instead, describe what *changes*.

* ✅ "She slowly turns her head and smiles, hair swaying gently in a breeze"
* ✅ "Leans forward slightly, eyes narrowing with a playful expression"
* ❌ "A woman standing on a beach in a red dress at sunset" — the image already shows this

**2. Be specific about intensity and pacing**

Use degree words to control how movements feel. The model is very responsive to pacing cues.

* ✅ "Slowly brushes hair behind her ear"
* ✅ "Gentle breeze lifts her hair, subtle movement"
* ✅ "Quickly glances over her shoulder"
* ❌ "Moves" — too vague, unpredictable result

**3. One camera movement per shot**

The model handles camera movement well, but most reliably when you give it one clear instruction. Combining multiple movements can cause jitter.

* ✅ "Slow dolly push-in"
* ✅ "Smooth pan to the right"
* ✅ "Camera holds still" — perfectly valid, keeps the focus on subject motion
* ❌ "Dolly in while panning left and tilting up" — too many movements at once

Use pacing words like "slow," "smooth," or "gentle" rather than technical parameters.

**4. Separate camera movement from subject movement**

This is the most common prompting mistake. Be explicit about what moves and what stays still.

* ✅ "She slowly turns her head to the right. Camera holds fixed framing."
* ✅ "Camera slowly pushes in. She holds her pose, only her hair moves in the wind."
* ❌ "Everything moving at once" — confuses the model

**5. Use sequential actions for complex scenes**

List actions in the order you want them to happen. The model follows temporal sequences well:

* "Looks down at her phone with a surprised expression, then looks up at the camera with excitement, gasps softly"
* "Takes a sip of coffee, sets the cup down, then glances out the window with a thoughtful expression"

**6. Add quality constraints at the end**

Append a short constraint line to any prompt for consistently better output:

* "No distortion. No jitter. Face stable, no deformation."

These instructions can help clarify your priorities, though they do not guarantee an artifact-free result.

#### Example Prompts

**Simple & effective:**

> She brushes hair behind her ear with a soft smile, gentle breeze, smooth and natural motion. No distortion.

**With camera movement:**

> Slow dolly push-in. She looks up toward the camera with a warm, inviting expression, lips parting slightly. Hair shifts gently. No jitter, face stable.

**Complex sequential action:**

> She glances down at her phone and her eyes widen with surprise, then she looks up at camera with excitement and laughs softly. Camera holds still. No distortion, no deformation.

<figure><img src="https://3286776735-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F9ACKOAU4nFWSXPeE2B4s%2Fuploads%2Fd5ckELPu976yYJvd4eNd%2Fimage.png?alt=media&amp;token=44dd8171-7831-43e5-ac7a-ed45d4a43c4b" alt="" width="375"><figcaption></figcaption></figure>

<figure><img src="https://3286776735-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F9ACKOAU4nFWSXPeE2B4s%2Fuploads%2FECA3q4hLKeYFBkYrkJVA%2Fimage.png?alt=media&amp;token=3fd7ed7f-f923-4517-bc10-288faf8be665" alt="" width="375"><figcaption></figcaption></figure>

#### Prompting With Audio

Audio is generated alongside the video on every model with audio enabled (Classic lets you turn it off), not as a separate step. The model reads the source image and your prompt to produce matching sound automatically.

**Three audio layers are generated simultaneously:**

* **Lip-synced speech** — Include short dialogue in your prompt with emotional context for best results (e.g., "She softly whispers 'Hey, I missed you'"). Keep dialogue to **5–10 words per line** for the cleanest lip sync.

<figure><img src="https://3286776735-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F9ACKOAU4nFWSXPeE2B4s%2Fuploads%2FUKUce5lVEzZREGbknS9E%2Fimage.png?alt=media&amp;token=fbd00dd6-cc69-4b30-b9bd-867ad2858963" alt="" width="375"><figcaption></figcaption></figure>

* **Sound effects** — Tied to the actions you describe. Footsteps, cloth movement, glass clinking, a sigh — the model generates these based on what's happening in the scene.

<figure><img src="https://3286776735-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F9ACKOAU4nFWSXPeE2B4s%2Fuploads%2Fe69bdLudDZaZhc3g03Yi%2Fimage.png?alt=media&amp;token=e53c99a3-4871-46ac-a0ca-4c8aecd35e71" alt="" width="375"><figcaption></figcaption></figure>

<figure><img src="https://3286776735-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F9ACKOAU4nFWSXPeE2B4s%2Fuploads%2F3fInD8yodex5K9ZGIjmG%2Fimage.png?alt=media&amp;token=85c07af0-0491-409f-942c-21a4cb9d6154" alt="" width="375"><figcaption></figcaption></figure>

* **Ambient sounds & music** — Background atmosphere inferred from the source image and prompt (city noise, ocean waves, quiet room tone, soft background music).

{% hint style="info" %}
**Audio tip:** Label emotions before dialogue for much better results. "She whispers softly" produces noticeably better lip sync and tone than just writing the dialogue alone.
{% endhint %}

#### Things to Avoid

| Don't do this                         | Why                                                                           | Do this instead                          |
| ------------------------------------- | ----------------------------------------------------------------------------- | ---------------------------------------- |
| Re-describing the image               | The model already sees the image — redundant prompting wastes space           | Focus on motion and change               |
| Multiple camera movements in one shot | Causes jitter and incoherent motion                                           | One clear camera movement per shot       |
| Contradicting the source image        | Creates visual confusion (e.g., describing a beach when the image is indoors) | Stay consistent with what's in the image |
| Overloading a short clip              | A short clip can't contain an entire story                                    | One clear moment per generation          |
| Vague motion ("she moves")            | Unpredictable results                                                         | Be specific about what moves and how     |
| Long dialogue lines                   | Lip sync degrades with long speech                                            | Keep dialogue to 5–10 words              |

{% hint style="info" %}
**Pro tip:** For clips longer than \~8 seconds, consider whether two shorter generations might give you better results. Quality stays highest in the 4–8 second range.
{% endhint %}

<figure><img src="https://3286776735-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F9ACKOAU4nFWSXPeE2B4s%2Fuploads%2Fyz0NYH9EpLINUew2upis%2Fimage.png?alt=media&amp;token=ef89775f-ba2a-423e-8053-0997c55f07bd" alt="" width="375"><figcaption></figcaption></figure>

### Requirements & Limitations

#### Subscription Requirements

* **FREE Tier**: ❌ Not available
* **PREMIUM Tier**: ✅ Available
* **ULTIMATE Tier**: ✅ Available

### Understanding Video Results

#### What to Expect

Cinematic handles complex motion, realistic physics, and multi-subject interactions very well. That said, AI video generation still has some limitations:

* **Hands & fingers** — Occasional extra or fused digits can appear in complex hand movements
* **Fine patterns** — Tight weaves, tiny checks, and detailed textures can shimmer or flicker. Simpler patterns work better.
* **Variation** — Results vary between generations, even with the same prompt
* **Iteration** — You may need to regenerate or tweak your prompt for the best result

### Troubleshooting Video Generation

#### Safety Check Messages

Video prompts, reference media, and finished videos all pass through our safety checks. If a check stops your video, the **Generation Failed** notice tells you which part was flagged and by which check:

* **Your video prompt was stopped by our safety check** (or **copyright check**). The prompt itself was flagged. Rewrite the flagged part and try again. Nothing is charged when a prompt is stopped before generation.
* **A reference image or video was rejected by our safety, copyright, or privacy check.** One of your reference photos or clips was flagged. The privacy check flags references that appear to show a real person; swap in a different reference.
* **The generated video was blocked by our safety check** (or **copyright check**). The model rejected the finished video. Try a different scene, or a different character design if the copyright check stopped it.
* **The safety checker is temporarily unavailable.** This is an outage on our side, not a problem with your prompt. Wait a moment and try again.

Moments for a video that fails after it starts generating are refunded automatically. Follow the notice to adjust the prompt or reference before trying again.

### Frequently Asked Questions

**Q: Which model should I pick?**

* **Cinematic** for the best quality (nudity included), **Everyday** to save Moments, **Intimate** for intimate scenes. See [Which Model Should I Use?](#which-model-should-i-use)

**Q: Should I use Intimate for all NSFW videos?**

* No. Cinematic handles teasing, undressing and nudity well, at better quality. Use Intimate for intimate scenes, or see [Pro Tip: Best Results for Intimate Scenes](#pro-tip-best-results-for-intimate-scenes) to combine the two.

**Q: What does Turbo do?**

* Turbo makes Intimate generate faster for a little more. The extra cost is shown beside Turbo in the model menu before you generate.

**Q: How much do videos cost?**

* Each model card shows its price for your current length and resolution, and the Generate button shows the final price. See [Example Prices](#example-prices).
* **Basic** also varies by resolution and duration; it is no longer a fixed-price clip.
* Adding reference photos or clips costs nothing extra.

**Q: Do I need a subscription?**

* Yes, video generation requires a PREMIUM or ULTIMATE subscription.

**Q: Can I use any image?**

* Yes, any generated image can be used as source for video.

**Q: What resolutions and lengths can I generate?**

* It depends on the model. See [All Models at a Glance](#all-models-at-a-glance).

**Q: How do I generate in 4K?**

* Pick **Cinematic** (or **Intimate** for intimate scenes) and choose **4K**.

**Q: How do I make a video longer than 15 seconds?**

* Only **Intimate** goes up to 30 seconds. Other models go up to 15 seconds (Classic: 12).

**Q: Where did Advanced, Standard and Adult go?**

* They were renamed: Advanced (Quality) is now **Cinematic**, Advanced (Fast) is **Cinematic Fast**, Standard is **Classic**, and Adult is **Basic**. In Studio, turn on Fast in the Cinematic row to select Cinematic Fast; find Classic under More models.

**Q: Can I use reference photos and clips, or set an end frame?**

* Everyday, Cinematic, Cinematic Fast, and Intimate support both photos and clips. Basic supports end frames and reference photos, but not clips. Use **Frames** for Start and End, or **References** for up to **5 photos total** and, on supported models, **3 clips** (15 seconds combined). Reference media adds no separate charge.

**Q: Can I add audio?**

* Audio is included for Everyday, Cinematic, Cinematic Fast, Intimate, and Basic. On Classic it's optional (turning it off halves the price).
* Audio supports speech lip sync, ambient sounds, action sounds, and background music.

**Q: Can I download videos?**

* Yes, you can download generated videos from the videos tab.

**Q: Can I generate multiple videos from same image?**

* Yes. Try different prompts, models, or settings. Wait for an in-progress video from that image to finish before submitting another.

**Q: What motion can I create?**

* Cinematic excels at complex, realistic motion — walking, turning, hair blowing in wind, clothing movement, facial micro-expressions, multi-subject interactions, and much more. It handles physics naturally, so don't be afraid to get creative.

**Q: How do I get the best results?**

* Start your prompt with the shot type, always describe the lighting, use one camera movement per shot, and add "No distortion. No jitter." at the end. See the [Video Prompting Guide](#video-prompting-guide) for detailed tips.

**Q: My video failed with a safety, copyright, or privacy message. Was I charged?**

* Prompts stopped before generation are not charged. If a video fails after it starts, its Moments are refunded automatically. The notice tells you whether the prompt, a reference, or the finished video was flagged. See [Safety Check Messages](#safety-check-messages).

***

Want to learn more? Check out [Image Generation](/create/image-generation.md) or explore [Custom Characters](/create/custom-characters.md)!
