Secrets AI Video Generator: How It Works, Quality, and Cost
Most AI companion platforms don't generate video. Secrets AI does — and it's the feature that most clearly differentiates the platform in a crowded market. Understanding what the video generator actually produces, what it costs in Moments, and when it's worth using is the key to getting value from this capability.
Independent reviewers rate the video output at 4.1/5 — competitive quality for AI-generated companion video, with natural movement and expressions in most outputs.
What Is the Secrets AI Video Generator?
The video generator converts static companion images into short animated video clips. You select a companion image, write a text prompt describing the desired action or movement, and the AI processes the request to produce a clip showing your companion in motion.
This feature is genuinely uncommon in the AI companion space. Comparing the major platforms:
- Character.AI: No video generation
- CrushOn AI: No video generation
- Janitor AI: No video generation
- Candy AI: Limited video functionality
- Replika: No video generation
The only competitors with comparable video capabilities are niche platforms like SweetDream AI and Xotic AI (which offers 4K 15-second clips). For a mainstream AI companion platform, Secrets AI's video generation is a meaningful differentiator.
Video generation requires Lite tier or above — it is not available on the free plan. See the free vs premium comparison for the full tier feature breakdown.
How Video Generation Works
The process is straightforward:
- Start with an image: Generate a new companion image in chat, or select an existing one from your character's image gallery. Higher-quality source images produce better video output.
- Write your prompt: Describe the movement, action, or expression you want. Specific prompts ("slowly turning her head and smiling") produce more coherent results than vague ones ("do something").
- Submit and wait: Processing takes approximately 2 minutes — longer than image generation but reasonable for the output.
- Review and save: The completed clip appears in your chat interface. Save it to your device if desired.
Videos reflect the character's appearance from the source image and adapt the movement to the conversational context if the prompt references the scenario.
Video Quality Assessment
The 4.1/5 rating reflects consistent but not flawless output:
Strengths:
- Natural character movement in most outputs
- Facial expressions are responsive to the prompt
- Character consistency between source image and video
- Smooth motion without obvious frame jumping in standard clips
Limitations:
- Quality degrades with very specific or complex prompt instructions
- Occasional glitches in hand or finger movement (common across AI video generation systems)
- Very short clips at lower tiers (3 seconds on Lite)
- Quality ceiling is lower than dedicated AI video platforms
Quality improves noticeably when using the Premium generation model, available on higher subscription tiers. Starting with clean, well-lit source images also makes a significant difference in output quality.
Moments Cost by Clip Length
This is the critical budget consideration. Video is the most Moments-intensive feature on the platform:
| Clip Type | Moments Cost |
|---|---|
| Short clip (3 seconds, Lite+) | ~50 Moments |
| Standard/longer clip | ~600 Moments |
For comparison, the same 600 Moments could alternatively generate:
- 12-24 images (at 25-50 Moments each), or
- 6 minutes of voice calls (at 100 Moments/minute), or
- 300-600 text messages (at 1-2 Moments each)
Video is expensive relative to other features. This matters for budget planning.
Monthly Video Budget by Tier
How many videos can you realistically create per month at each subscription level?
| Tier | Monthly Moments | Short Clips (50 Mo) | Long Clips (600 Mo) |
|---|---|---|---|
| Lite | 1,000 | ~20 | ~1-2 |
| Plus | 3,000 | ~60 | ~5 |
| Premium | 8,000 | ~160 | ~13 |
| Ultimate | 15,000 | ~300 | ~25 |
These are pure-video calculations — in practice, you'll also be spending Moments on images and occasionally voice, so real video output per month will be lower.
Practical example for Premium users (8,000 Moments): A mixed-use month — 100 images (3,500 Moments) + 5 long videos (3,000 Moments) + 15 minutes voice (1,500 Moments) = 8,000 Moments exactly. That's the ceiling of a full Premium month with diverse media use.
For Moments pricing and bulk purchase options, see the complete pricing guide.
Video vs Images vs Voice — When to Use Each
| Feature | Cost | Best For |
|---|---|---|
| Image | 25-50 Moments | Character visualization, scene setting |
| Short video (3s) | ~50 Moments | Quick motion preview, testing prompts |
| Full video | ~600 Moments | Significant moments, sharing, saving |
| Voice call | 100/min | Real-time conversational immersion |
| Text | 1-2/message | Daily conversation, roleplay |
Strategy recommendation: Generate images first to establish how your companion looks in a particular scenario. When you have a strong source image you're happy with, then spend on video generation. This avoids spending 600 Moments on a video from a suboptimal source image.
Tips for Better Video Results
From practical testing:
- Use high-quality source images. The video inherits both the strengths and weaknesses of the source image. A blurry or glitched source produces a blurry or glitched video.
- Write specific, action-focused prompts. "Slowly smiling and leaning forward" produces better results than "look friendly." Describe physical movements rather than emotional states.
- Start with short clips. Test a prompt concept with a 50-Moment short clip before committing 600 Moments to a full-length version.
- Use the Premium generation model. The quality difference is noticeable for video — if you're on Premium or Ultimate, make sure you're selecting the advanced model.
- Match prompts to source image context. A video prompt asking for beach movement from a formal portrait creates coherence problems. Source image and prompt should align.
Who Should Use the Video Generator?
Worth investing Moments in if:
- You value visual content alongside conversation
- You want to capture and save significant moments from your companion relationship
- You're on Premium or Ultimate with sufficient Moments to use it regularly without depletion anxiety
Better to skip or minimize if:
- You're primarily a text conversation user
- You're on Lite or Plus with 1,000-3,000 Moments monthly — video generation will crowd out other media use
- You're budget-conscious and want to maximize interaction time within your Moments allocation
Tier recommendation: Ultimate ($39.99/month) for heavy video creation (15,000 Moments supports 25 long clips or 300 short clips monthly). Premium ($19.99/month) for moderate video use alongside image generation and voice.
See the full platform review for the overall assessment including all feature ratings.
Try video generation on Secrets AI →
FAQ
Video length depends on your subscription tier and the specific generation option selected. Lite tier produces short 3-second clips. Plus, Premium, and Ultimate tiers unlock longer clip generation. The exact maximum length for longer clips is not publicly specified, but costs scaling to 600 Moments per clip indicate meaningful length differences between short and full clips.
No. Video generation requires Lite tier ($5.99/month) or above. The free plan provides 200 starting Moments but no video access — those Moments can only be spent on images and voice calls.
It depends on your tier and clip length. On Premium (8,000 Moments), you can make approximately 13 full-length clips or 160 short clips — but only if you spend your entire Moments allocation on video. In practice, most users spend Moments on a mix of images, video, and voice, so actual video output is lower. Ultimate tier (15,000 Moments) gives the most video headroom.
The output quality is rated 4.1/5 by independent reviewers. Videos show natural character movement and responsive facial expressions in most outputs. Quality is strong for AI-generated companion video — comparable to Candy AI's limited video implementation and better than most alternative platforms that offer any video at all. Quality varies with source image quality and prompt specificity; occasional glitches in hands or peripheral elements occur.