Secrets AI Video Generator: How It Works, Quality, and Cost
Video generation from AI companion images is the feature that most distinguishes Secrets AI from its direct competitors. Character.AI doesn't offer it. CrushOn AI doesn't offer it. Janitor AI doesn't offer it. Candy AI has limited video capabilities but nothing as developed as Secrets AI's system. If you're evaluating whether video is worth the Moments cost, this guide covers the mechanics, quality assessment, and the math by tier.
What the Video Generator Actually Does
The Secrets AI video generator takes a still image of your AI companion and animates it based on a text prompt you provide. The output is a short video clip showing your companion moving, expressing, or performing the action described in the prompt.
This is categorically different from the kind of AI-generated video tools used for general media production. It's specifically designed around companion images — the character's appearance, personality context, and established relationship scenario carry through into the video output.
The feature is available on Lite tier and above — not on the free plan. It represents one of the clearest functional reasons to upgrade beyond free.
The Generation Process
The four steps are straightforward:
- Generate or select a companion image. Either create a new image from the chat interface (25–50 Moments) or use an existing image from your character's gallery. Image quality directly affects video quality — start with your best images.
- Write a text prompt. Describe the movement, expression, or action you want. Specific prompts produce better results than vague ones. "Smiling and turning her head slowly" works better than "look at me."
- Wait for generation. Processing takes approximately 2 minutes per clip. This is consistent across tiers — the generation time doesn't change based on your plan.
- View and save the clip. The completed video can be viewed in-platform and saved. Review before spending Moments on similar clips if the result doesn't match expectations.
Video clips range from 3-second shorts (available on Lite) to longer clips available on Plus and above. The prompt complexity and clip length both affect output quality.
Quality Assessment
Reviewers rate Secrets AI video at 4.1/5 — the second-highest feature rating on the platform after chat (4.4/5). The specific quality notes from testing:
- Videos "look good and move smoothly most of the time"
- Realistic character movement and facial expressions in most outputs
- Natural motion without the jerky or distorted movement common in earlier AI video tools
- Occasional quality variations depending on prompt complexity — simple movement prompts produce more consistent results than complex action sequences
- Quality improves when using the Premium generation model (available on Premium tier and above)
The main quality caveat is consistency — like all AI video generation, outputs vary. Some prompts produce excellent results on the first try; others require iteration. This is why starting with shorter, cheaper clips before committing Moments to longer videos is practical advice.
Moments Cost by Clip Type
| Clip Type | Moments Cost |
|---|---|
| Short (3-second) clip | ~50 Moments |
| Full-length video clip | ~600 Moments |
The gap between short and long clips is significant — a full clip costs 12x what a short clip costs. For initial testing and lower-stakes use, short clips are the economical choice.
Video Budget by Subscription Tier
| Tier | Monthly Moments | Short Videos (3s) | Full-Length Videos |
|---|---|---|---|
| Lite | 1,000 | ~20/month | ~2/month |
| Plus | 3,000 | ~60/month | ~5/month |
| Premium | 8,000 | ~160/month | ~13/month |
| Ultimate | 15,000 | ~300/month | ~25/month |
These numbers assume Moments are spent exclusively on video — in practice, you'll also use Moments on images, text messages, and possibly voice calls. The free vs premium breakdown shows the full Moments math across all feature types.
How Video Compares to Other Media Costs
For the same Moments budget, here's what you can generate:
| Feature | 600 Moments Produces |
|---|---|
| Full video clip | 1 clip |
| Short video clips (3s) | ~12 clips |
| Images | 12–24 images |
| Voice calls | 6 minutes |
| Text messages | 300–600 messages |
This makes video the highest cost-per-unit feature on the platform. Users who value video content should ensure they're on Premium or Ultimate where the Moments allocation is large enough to include video without depleting the entire monthly budget.
Tips for Better Video Results
Use high-quality source images. The video inherits the quality of its source image. Blurry or anatomically inconsistent images produce worse video output. Generate a few image variants and select the sharpest one before converting to video.
Keep prompts specific but not overly complex. "Walking slowly and smiling" works reliably. "Dancing a complex choreography with multiple wardrobe changes" is too complex for current AI video capability and will likely produce unsatisfying results.
Test short clips first. Before spending 600 Moments on a full-length video, generate a 3-second clip (~50 Moments) with the same prompt to verify the motion quality meets your expectations. Iterate on the prompt at low cost before committing to the full version.
Use the Premium generation model. If you're on Premium or Ultimate tier, the advanced generation model produces noticeably better video quality. Select it from the generation options rather than defaulting to the standard model.
Save your best source images. Characters with consistent, high-quality reference images in their gallery produce more stable video outputs. The more detail the model has to work with, the more consistent the motion and expression rendering.
Who Gets the Most Value from Video
Worth using regularly if:
- You value visual companion content alongside text interaction
- You want to save and revisit media from your companion relationship
- You're on Premium ($19.99) or Ultimate ($39.99) where Moments cover regular video use
Not worth prioritizing if:
- You're on Lite (1,000 Moments) — full videos cost too much relative to your monthly budget
- You're primarily a text-based user and don't care about visual media
- Moments management is already a concern on your current plan
How Competitors Compare on Video
The video generation gap between Secrets AI and competitors is one of the clearest differentiation points in the category:
| Platform | Video Generation |
|---|---|
| Secrets AI | Yes — clips from companion images, 50-600 Moments |
| Candy AI | Limited video capabilities |
| CrushOn AI | No |
| Janitor AI | No |
| Character.AI | No |
| GirlfriendGPT | No |
| Kalon AI | No |
For users where video is a meaningful part of the value proposition, there is currently no direct replacement that offers the same companion-integrated experience. The Secrets AI alternatives page covers the broader competitive landscape.
The full feature context — including how video fits into the platform's chat and image ecosystem — is in the complete review.
Frequently Asked Questions
Short clips run approximately 3 seconds and cost ~50 Moments. Full-length video clips are longer and cost approximately 600 Moments each. The Lite tier only supports 3-second clips; Plus and above can generate full-length versions. Generation time for all clips is approximately 2 minutes.
No. Video generation requires a Lite tier subscription or above. Free accounts can access the character library and text chat but cannot generate images or video. The 200 starting Moments on free accounts are not sufficient for video even if the feature were unlocked — a full clip costs 600 Moments.
It depends on your tier and how you allocate your Moments. On Plus (3,000 Moments), you can generate approximately 5 full-length clips per month if Moments are used exclusively on video, or ~60 short (3-second) clips. On Premium (8,000 Moments), approximately 13 full-length clips or ~160 short clips. On Ultimate (15,000 Moments), approximately 25 full-length clips or ~300 short clips. In practice, you'll mix video with images and chat, so actual video output will be lower than these maximums.
Reviewers rate the video quality at 4.1/5. The videos show realistic character movement and expressions in most cases, with smooth motion that avoids the obvious distortion of earlier AI video tools. Quality varies by prompt complexity — simple movement prompts produce the most consistent results. Complex action prompts or unusual angles occasionally show quality inconsistencies. Using the Premium generation model (available on Premium tier) and high-quality source images improves output consistency.