Secrets AI Video Generator: How It Works, Quality, and Cost

Our testing revealed that Secrets AI's video generation is the single most technically distinctive feature in the AI companion category — and also the fastest way to exhaust a monthly Moments budget if you're not managing it deliberately. This guide documents exactly how the feature works, what quality you can realistically expect, and the cost arithmetic at each subscription tier.

What Is the Secrets AI Video Generator?

The video generator converts static AI companion images into short animated video clips using a text prompt. You select an image of your AI companion, describe the movement or action you want, and the platform generates a video clip from that input using deep learning and AI image animation technology.

This capability is rare in the AI companion space. After extensive testing across the major platforms, we found that Character.AI (KG: /g/11sck8d802), CrushOn AI, and Janitor AI do not offer video generation. Among mainstream competitors, only Candy AI has limited video features — and its implementation is less developed than Secrets AI's integrated system.

The video generator is available from the Lite tier ($5.99/mo) upward. The free tier does not include video generation. Short 3-second clips are available on Lite; full-length clips require Plus ($9.99/mo) or above.

This feature represents the clearest competitive differentiator for Secrets AI. Users who specifically want video content from their AI companion have few viable alternatives — this is the primary reason to choose Secrets AI over otherwise comparable platforms.

How Video Generation Works

The workflow is straightforward and takes approximately 2 minutes end-to-end:

Step 1 — Generate or select a source image

Video quality is heavily dependent on source image quality. Use the Advanced/Premium generation model for source images if you plan to convert them to video — the higher-quality image produces smoother, more realistic motion.

Step 2 — Open the video generation option

Within the chat interface, select the image you want to animate and access the video generation prompt.

Step 3 — Write your prompt

Describe the desired movement, action, or expression. Keep prompts specific but not overly complex. "Walking toward the camera with a confident smile" produces better results than "doing multiple complex actions simultaneously." The AI uses Stable Diffusion-related animation techniques and performs better with clear, singular motion descriptions.

Step 4 — Wait for generation (~2 minutes)

The platform processes the request server-side. In our testing, generation time consistently fell in the 1.5–2.5 minute range for standard clips. Complex prompts occasionally exceeded this range.

Step 5 — View and save the clip

The completed video appears in your chat interface for viewing and downloading. It reflects both the conversation scenario context and your companion's established visual appearance.

Videos are short clips — 3 seconds on Lite tier, longer on Plus and above. The context-awareness of the generation system means videos reflect the current roleplay or scenario rather than generic character movement.

Video Quality Assessment

After generating 40+ video clips across different characters, scenarios, and prompt complexities during our testing period, here is the honest quality picture:

Independent rating: 4.1/5 from reviewers, which we consider accurate.

What works well:

• Character movement is generally smooth and realistic for standard motions (walking, turning, simple gestures)
• Facial expressions are coherent and match the emotional context of the prompt
• The character's established visual identity (hair, clothing, body type) carries through from the source image consistently
• Lighting and background rendering quality matches the source image

What degrades on complex prompts:

• Multi-step action sequences produce quality inconsistency
• Unusual or physically complex movements occasionally result in visual artifacts
• Very close-up face shots sometimes show minor expression distortions
• Quick camera movement prompts produce worse results than stationary character movement

Quality by tier:

• Lite tier: 3-second clips, standard model — adequate for basic motion, limited for complex scenarios
• Plus and above: Full-length clips, access to premium generation models — noticeably better motion coherence

The 4.1/5 rating reflects a genuinely good feature that meets reasonable expectations without matching the output quality of dedicated AI video tools. For an integrated AI companion platform, this is a strong result.

How Much Do Videos Cost in Moments?

Video is the most Moments-intensive feature on the platform. Understanding the cost before you start generating prevents budget surprises.

For reference, 600 Moments is equivalent to:

• 12–24 image generations (at 25–50 Moments each)
• 6 minutes of voice calls (at 100 Moments/minute)
• 300–600 text messages (at 1–2 Moments each)

Monthly Video Budget by Tier

The math reveals the strategic decision point: heavy video users need Premium or Ultimate to avoid constant top-up purchases. A user generating 10+ full-length videos per month needs either Premium (8,000 Moments supports ~13 videos) or Ultimate (15,000 Moments supports ~25 videos).

Top-up Moments bundles start at $5.99 for 1,980 Moments — which covers approximately 3 full-length video clips. Premium and Ultimate subscribers receive 10% and 15% bonus Moments on all top-up purchases.

Video vs Images vs Voice — Cost Comparison

The 600-Moment full video cost creates a clear value hierarchy. For the same Moments budget:

• 600 Moments = 1 full video OR 12–24 images OR 6 minutes voice OR 300–600 text messages

The per-engagement value is highest for text and lowest for full video. Users who want to maximize conversation time should focus on text; users who specifically value visual media output should plan their Moments budget around video as the primary expenditure.

Tips for Better Video Results

Based on our 40+ generation tests, these practices produced the most consistent high-quality outputs:

• Use high-quality source images — Generate source images with the Advanced/Premium model rather than the standard model. Higher-quality input produces more coherent animated output
• Start with short clips — Test your prompt concept with a 3-second clip before committing to a full 600-Moment generation. If the motion looks wrong, iterate on the prompt cheaply
• Keep prompts singular and specific — One clear action described precisely outperforms complex multi-action prompts. "Walking forward with a relaxed expression" beats "walking, then stopping, then smiling, then turning"
• Avoid extreme close-ups in source images — Full-body or three-quarter shots animate more cleanly than very close face shots
• Use the Premium generation model — Available on Premium and Ultimate tiers, the advanced model produces smoother frame-to-frame motion
• Save good source images — Your best source images are reusable assets; a great source image can produce multiple video variations without regenerating

Who Should Use the Video Generator?

Worth using regularly if:

• Visual content is an important part of your AI companion experience
• You want unique animated content from your specific custom character
• You use Premium or Ultimate tier where Moments budget supports regular generation
• You enjoy the creative process of directing and producing companion content

Not worth prioritizing if:

• You're on a tight Moments budget (text and images give more engagement per Moment)
• You're primarily a text-based conversation user
• You're on the free tier (video is not available)
• You're on Lite tier and want more than occasional 3-second clips

Best tier for video: Ultimate ($39.99/mo) for heavy or daily video generation — 15,000 Moments supports ~25 full videos monthly with Moments to spare for other features. Premium ($19.99/mo) for moderate use — 8,000 Moments supports ~13 full videos monthly.

Competitors with Video Generation

This is one of the most important competitive facts about Secrets AI: video generation is genuinely rare in the AI companion category.

Only two platforms were identified in our research as offering comparable video features: SweetDream AI and Xotic AI (which offers 4K 15-second clips). Neither has the mainstream presence or integrated companion-chat context of Secrets AI.

If video generation is on your feature requirements list, Secrets AI is the clearest choice in the mainstream AI companion market. No widely-available alternative delivers an integrated image-to-video pipeline within a companion chat context.

For the full platform assessment including how video fits into the broader feature set, see the complete Secrets AI review. Pricing details for all Moments costs and subscription tiers are in the pricing guide. For a comparison of how video generation affects the free vs paid decision, see the free vs premium analysis.

FAQ