Question 1

How does DeepSwapAI pricing work?

Accepted Answer

DeepSwapAI uses a credit-based pay-as-you-go model with three published tiers: Starter Pack at $4.99, Popular Pack at $9.99, and Pro Pack at $19.99. Credits work across every feature on the platform — image face swap, video face swap up to 10 minutes, character animation, lip sync, and batch processing through the REST API. Credits never expire and there is no recurring subscription. For agencies and enterprise customers needing volume discounts, dedicated SLA, and invoice billing, contact partner@deepswapai.com for custom enterprise pricing.

Question 2

Is my data safe?

Accepted Answer

Yes. DeepSwapAI is architected with privacy-by-design and complies with GDPR Article 5(1)(c) data minimization, CCPA, and the United States TAKE IT DOWN Act 2025. All uploads are auto-deleted within 24 hours on standard tier and within 1 hour on Enterprise tier. We use AES-256 encryption at rest and TLS 1.3 in transit. We do not train, fine-tune, or extract biometric templates from user uploads. For full audit details and DPO contact, see /trust and /privacy.

Question 3

What file formats are supported?

Accepted Answer

For images: JPG, PNG, WebP, and animated GIF. For videos: MP4, MOV, AVI, and WebM with H.264, H.265, VP9, or AV1 codec support. For audio companion features: MP3, WAV, FLAC, OGG. Output format defaults to input but can be converted to any supported format. Color spaces include sRGB, DCI-P3, and Rec. 709. Subsampling 4:2:0 and 4:4:4 both work. RAW camera formats are not currently supported — convert to JPG/PNG first if working with photographer source material.

Question 4

How long does processing take?

Accepted Answer

Image face swaps complete in 5-15 seconds at 1080p, 15-25 seconds at 4K. Video face swaps process at approximately 1-2 seconds of inference per second of input video — a 1-minute 1080p video typically takes 60-120 seconds. Lip sync runs faster at 0.5 seconds per second of input. We use dedicated NVIDIA A100 and H100 GPU clusters with typically zero queue delays during normal operating hours. Enterprise users get reserved GPU capacity with sub-second image latency SLAs.

Question 5

Can I use results commercially?

Accepted Answer

Yes, you own the rights to face-swap output you generate, provided you have legal rights to the source images and target media. Our Terms of Service formally grant commercial usage rights to all output. Enterprise customers receive an indemnification clause in their contract for additional coverage. For agencies generating content for client campaigns, no additional licensing is required from us. However, you remain responsible for ensuring you have model releases or consent for any depicted individuals — see our /ethics-policy for detailed obligations.

Question 6

Do you offer API access?

Accepted Answer

Yes. DeepSwapAI exposes a complete REST API at https://deepswapai.com/doc for programmatic access to every feature. Authentication uses bearer tokens issued in your account dashboard. Rate limits scale by tier: 100 req/month on Starter, 1K on Popular, 5K on Pro, custom on Enterprise (typically 100K+/month). Webhooks are available for long-running operations. SDKs are published for Python, Node.js, and TypeScript. The API exposes a /batch endpoint accepting up to 50 files per call. Documentation includes a Postman collection and curl examples for every endpoint.

Question 7

How does the credit system work?

Accepted Answer

DeepSwapAI uses a credit-based pay-as-you-go model with no subscription required. Credits are consumed by feature: image face swap costs 1 credit per output, video face swap costs 5 credits per second of input video, character animation 10 credits per second, lip sync 3 credits per second. Credits never expire and are reusable across every feature. New accounts receive 5 free credits to test the platform. Volume discounts of 20–40% apply at Pro Pack ($19.99) and enterprise tiers.

Question 8

What video resolutions are supported?

Accepted Answer

DeepSwapAI processes input video up to 4K (3840×2160) and outputs at the same resolution as input. Internal pipeline supports 480p, 720p, 1080p, 2K, and 4K with no quality degradation across the chain. Frame rate up to 60 fps is preserved end-to-end; variable frame rate (VFR) input is normalized to constant frame rate (CFR) before swap. For 4K cinema-quality output we recommend Pro Pack credits which run on dedicated NVIDIA H100 instances for sub-second-per-frame inference.

Question 9

Do you support batch processing?

Accepted Answer

Yes. The /batch endpoint and UI both accept up to 50 files per single request, processing in parallel on the GPU pool. Use cases include: e-commerce product catalog face-swaps for ad localization, agency client deliverables (50 hero shots × 5 face variations), and creator workflows (TikTok templates × creator face). Batch jobs return a single ZIP archive plus a manifest with per-file status. Webhook callbacks notify completion for long jobs. Pro Pack and Enterprise tiers raise the per-batch limit to 200 files.

Question 10

What is character animation?

Accepted Answer

Character animation puts your face onto a target video while transferring the target's body movements, head pose, and lip sync to your face. Built on Wan 2.2 + temporal coherence tracking, it is the right tool for: putting yourself in dance videos, creating creator-style talking-head clips from a single photo, or producing animated avatar performances. Different from straight face-swap which preserves the original actor's identity in everything except the face — character animation makes the output 'be' you in motion and expression too.

Question 11

Is there a public API?

Accepted Answer

Yes. DeepSwapAI's full REST API is documented at /doc with bearer-token authentication. Rate limits scale by tier: 100 req/month on Starter, 1K on Popular, 5K on Pro, custom on Enterprise (typically 100K+/month). Webhooks support long-running operations. SDKs are published for Python, Node.js, and TypeScript. The API exposes /face-swap, /video-swap, /character-animate, /lip-sync, /batch, /text-to-video, and /enhance. Includes Postman collection and curl examples per endpoint.

Question 12

How do I integrate DeepSwapAI into my product?

Accepted Answer

Three integration paths: (1) REST API direct — embed our endpoints into your backend, suitable for SaaS products. Bearer token auth, JSON I/O. (2) Iframe embedding — drop our hosted UI into your site via /embed, useful for content publishers. Includes JS callback events for the full lifecycle. (3) White-label — enterprise tier provides a fully branded subdomain with our infrastructure and your branding. Email partner@deepswapai.com for white-label SLAs and custom-model fine-tuning.

Question 13

Is data shared with third parties?

Accepted Answer

No. DeepSwapAI processes user uploads exclusively on our own infrastructure (NVIDIA A100/H100 GPU pools we operate) and does not forward biometric data to third-party providers. We use Cloudflare for CDN and Stripe for payments — neither receives uploaded media. We do not sell, rent, or share user data with advertisers or analytics brokers. Aggregate, fully anonymized usage statistics are published in our annual transparency report. See /privacy for full data flow disclosure.

Question 14

What about deepfake misuse?

Accepted Answer

DeepSwapAI takes synthetic-media abuse seriously. We embed C2PA Content Credentials in every export so platforms (TikTok, Instagram, YouTube, X) can automatically display 'AI-generated' badges per EU AI Act Article 50. Our upload classifier blocks known politicians, minors, and non-consenting public figures. The /takedown channel accepts non-consensual imagery reports per US TAKE IT DOWN Act 2025 and processes them within 48 hours. We cooperate with NCMEC and law enforcement on CSAM cases.

Question 15

Do you offer enterprise SLAs?

Accepted Answer

Yes. Enterprise contracts include: 99.9% uptime SLA with credits for breach, dedicated GPU pool with sub-second image inference latency, 24/7 priority support with named CSM, custom rate limits (typically 100K+ req/month), volume pricing, white-glove onboarding, indemnification clause for IP and synthetic-media compliance, and quarterly security audits sharing. Typical enterprise customers are agencies, SaaS platforms, and content studios processing 10M+ creative assets/month. Contact partner@deepswapai.com for terms.

Question 16

What models do you use?

Accepted Answer

Our pipeline combines best-in-class open-source models: RetinaFace (face detection), HRNet (68-point landmark alignment), ArcFace and AdaFace (identity encoding), InsightFace inswapper-128 (primary GAN-based swap), HiFiVFS (diffusion-based refinement for premium tier), Wav2Lip and MuseTalk (lip sync), Wan 2.2 (character animation). All models run on TensorRT-compiled engines on NVIDIA A100 (Starter/Popular) or H100 (Pro/Enterprise) GPUs for 2.6× throughput on Pro/Enterprise tiers.

Question 17

How do I cancel a subscription?

Accepted Answer

There is no subscription to cancel — DeepSwapAI uses one-time credit packs only, no recurring charges. Once you purchase a pack, the credits are yours indefinitely with no expiration. If you wish to delete your account entirely, go to /account/settings and click 'Delete Account' which permanently purges your account, history, and any remaining credits within 30 days per GDPR Article 17 right to erasure. Refunds for unused credits within 14 days of purchase per EU consumer law are processed via the original payment method.

Question 18

Is DeepSwapAI affiliated with deepswap.ai?

Accepted Answer

No. DeepSwapAI (deepswapai.com — this site) is an independent professional AI face morphing studio. We are NOT affiliated with, owned by, or related to deepswap.ai which is a separate, unrelated company. We are the sister brand of FaceSwapAI (faceswapai.com), the consumer-tier free platform. Both DeepSwapAI and FaceSwapAI are operated by the same team. Per EU consumer protection law and US trademark norms, our schema.org Organization includes a `differentFrom` property specifically disambiguating the two.

常見問題解答

🚀 Getting Started

🔒 Privacy & Security

⚙️ Technical

💳 Pricing & Plans

🔗 API & Enterprise

Still have questions?