Deepfake Maker
Deepfake Maker is a browser-based AI face swap tool for photos, videos and GIFs, bundled with AI image and video generation and editing. A free tier works without installation or sign-up, while advanced functions run on credits.
What is Deepfake Maker?
Deepfake Maker is a browser-based face swap and deepfake tool: nothing is installed and the swap interface sits directly on the home page. Three families are offered, photo, video and GIF, each with variants for a single face, multiple faces, batch mode of up to 10 images, character swap, head swap and gender swap. Accepted formats are PNG, JPG, JPEG and WEBP for images, MP4, MOV and GIF for video. Before running a job you choose between two models, Basic, described as faster and cheaper, and HD, described as sharper, with an optional HD Enhance toggle. On video, faces are detected automatically, the clip length is locked once detection has run, and the interface prices that detection at 2 credits. Around the swap sits a much wider catalogue. Image generation covers text to image, image to image, where style, light and mood change while the structure is kept, and generation guided by 1 to 5 reference images. Image editing covers old photo restoration and colourisation, background removal, object removal or addition from a description, and upscaling. Video generation covers text to video, image to video and video to video, while video editing adds enhancement, upscaling and watermark removal. The footer leads to further generators: avatar, thumbnail, anime photo, talking photo, age and haircut filters, and effect generators such as kissing, twerk, bikini, jiggle and dance. Several third-party models get their own dedicated pages, among them GPT Image 2, Nano Banana AI, Pro and 2, Gemini 3.1 Flash Image, Seedance and Seedance 2.0. The site claims a 2 to 5 second turnaround per submission, free use without sign-up and without a watermark, credits for the advanced functions, and server-side processing with no retention of uploaded media. The publisher is thinner on detail. The privacy policy describes a company based in Hong Kong without naming it, while the terms name an entity called VideoFaceSwap Inc. The domain was registered on 30 August 2024, the first Wayback capture dates from 3 September 2024, and the footer reads Deepfake Maker 2024. A JSON-LD block on the home page self-declares a 4.9 rating over 89,868 reviews, a figure published by the site itself and impossible to verify.
What it does
- Swap a face in a photo, a video or a GIF
- Handle several faces at once, or up to 10 images in batch mode
- Generate an image from text, from a source image, or from 1 to 5 reference images
- Animate a still photo into a video, or restyle an existing video
- Restore and colourise an old photo, cut out the background, erase or add an object, upscale the resolution
- Remove a watermark from a video and upscale it
- Apply face filters: age, haircut, body shape and gender swap
When to use Deepfake Maker / When not to
A quick filter to help you decide if Deepfake Maker is the right fit.
When to use Deepfake Maker
- Social media creators making memes, reaction clips, shorts and animated GIFs, who need a shareable result in seconds without opening a video editor
- Marketers running visual experiments: testing campaign faces and creative ideas before committing to a shoot
- Independent filmmakers and writers previewing characters and building character teasers for a story
- Freelance video editors who need a quick swap, upscale or watermark removal outside their usual editing suite
- Complete beginners, since the product pages repeat that no editing skills are required and the workflow is three steps
When not to use Deepfake Maker
- Anyone who cannot obtain the consent of the people whose faces are used: the home page FAQ places that obligation entirely on the user, and impersonation, misleading or harmful content are forbidden
- Anyone counting on publishing the output: the terms restrict the service to personal study and academic research and forbid sharing generated content without permission, which contradicts the product's own social sharing pitch
- Journalists, investigators and anyone needing evidentiary value, since no watermarking and no provenance signature is described anywhere
- Procurement-led buyers who need a legal identity matching the brand, a postal address, a named subprocessor list or a certification: none of that is published
- Minors and anyone building on top of the service: the minimum age is 13, raised to 16 in the EEA, and there is no public API and no developer documentation to automate anything
How to use Deepfake Maker
A typical end-to-end flow, from setup to results.
- Open the home page: the tool is embedded there, and the free use claimed by the site needs no account
- Pick the tab matching your source, PHOTO FACE SWAP or VIDEO/GIF FACE SWAP
- Pick the sub-tab: Single Face, Multiple Faces or Batch Mode, the latter taking up to 10 source photos against one target face
- Upload the original photo or video, then add a clear photo of the target face
- On video, run Face Detection first: the clip length is locked once detection is done, and each detection is priced at 2 credits
- Choose the Basic or HD model, and switch HD Enhance on for a sharper result
- Start the swap and wait for the render, announced at 2 to 5 seconds per submission
- Download or share the result: right-click on desktop, long press on mobile to save it to the gallery
- Use the History button to find earlier generations
- Sign in and claim or buy credits when you need the advanced features, the higher limits or the history
Pros & Cons
Pros
- Free use is claimed with no sign-up and no watermark, alongside a permanent free tier on the pricing grid
- Unusually broad catalogue gathered in one place: swapping, image generation, video generation, editing, filters and effects
- The tool is embedded on the home page and usable straight away, with no installation, from a desktop, tablet or mobile browser, and a render announced at 2 to 5 seconds
- Batch mode up to 10 images and native handling of multiple faces in a single pass
- Credits are advertised as never expiring, and packages are described as one-time payments rather than subscriptions
- Recent third-party models are exposed explicitly, among them GPT Image 2, Gemini 3.1 Flash Image, Seedance 2.0 and Nano Banana
- Written commitment not to train models on user data, with uploaded media announced as deleted within 24 hours, and a refund policy that spells out its 7-day exceptions
Cons
- No price is visible: the five pricing cards are filled in client-side, so nothing is readable without JavaScript or an account, and neither the FAQ nor the refund policy carries a figure
- The terms name VideoFaceSwap Inc and refer to VideoFaceSwap throughout, an entity different from the site's brand, and no postal address, registration number or named director appears anywhere
- The documents contradict each other: auto-renewing monthly or annual subscriptions in the terms against one-off credit purchases on the pricing and refund pages, and never storing media on the home page against deletion within 24 hours in the privacy policy
- The privacy policy rules out advertising and profiling, then describes ad network partners using cookies and pixels for targeted advertising
- Three different vintages of legal documents: terms updated 1 September 2024, privacy policy 1 July 2025, refund policy 1 January 2026
- No API and no developer documentation, no named subprocessor list, no certification claimed such as SOC 2 or ISO 27001, no EU Article 27 representative and no DPO, and no GDPR compliance claim, the framework invoked being Hong Kong's
- A single email address covers support, legal and privacy and is obfuscated in the HTML, there is no company LinkedIn page, and the self-declared 4.9 rating over 89,868 reviews and the named testimonials cannot be verified
Pricing & Plans
A permanent free tier exists: the pricing page sets a "Free User, basic features for everyone" column against a "Paid User, advanced features for creators" column, and the home page FAQ claims free use without signing in and without a watermark. Paid usage runs on credits, the spending unit of every generation or edit. Consumption varies with the function, counted per generated image, per second of video or per GIF frame; the balance is debited only when a job starts, and an estimate is displayed before confirmation. One cost is stated explicitly in the interface: 2 credits per face detection on video. The pricing page states that all packages are a one-time payment with no expiration, and top-ups can be added without changing plan, stacking on the balance and being consumed after the included credits. No amount is published anywhere. The five pricing cards are client-side skeletons, and neither the served HTML, nor the underlying page payload, nor the ten-question pricing FAQ, nor the refund policy carries a single figure, so the lowest price point cannot be stated here. Accepted payment methods are credit cards, debit cards and Alipay. Credit purchases are in principle final, with exceptions within 7 days for a double charge, a major outage attributable to the publisher or an unauthorised payment, and an announced response time of 3 to 7 business days. The terms separately describe auto-renewing monthly or annual subscriptions, again without any amount.
- the free column of the pricing grid
- matching the home page claim of free use without signing in and without a watermark
- the paid column
- flagged Recommended
- unlocking the advanced feature set
- Five credit packs are laid out on the desktop pricing grid
- but no pack name and no amount is readable in the page served by the site
- The comparison grid runs on ten lines
- Credits Never Expire
- Basic Face Swap Model
- Advanced Face Swap Model
- Multi Face Swap Video
- Multi Face Swap Photo
- Batch Image Face Swap
- AI Image Generator
- no daily limit
- no duration or size limit
- HD enhance result
- more credits
- priority queues
- higher maximum resolution and duration
- priority rendering
- advanced features
- and higher per-task and concurrency caps
Data, GDPR & hosting
A consolidated view of how Deepfake Maker handles your data.
GDPR overview
There is no explicit claim of GDPR compliance anywhere on the pages retrieved, and no denial either. The Privacy Compliance Statement invokes the Personal Data (Privacy) Ordinance (Cap. 486, Laws of Hong Kong) rather than the GDPR, and the terms name Hong Kong law as the governing law. GDPR-style mechanisms are nevertheless cited: Standard Contractual Clauses for transfers from the EEA, and data processing agreements with all third-party providers. The minimum age is 13, raised to 16 for EEA residents. Users may request deletion of their personal data at any time by writing to the publisher, and data breaches are said to be notified to users and regulators as required by law. No EU representative under Article 27 and no data protection officer is named. The privacy policy carries a last-update and effective date of 1 July 2025.
Who owns the data?
The terms never state in so many words that the user owns the generated outputs. What they do state is responsibility: users accept full responsibility for any content they create, upload or share, must respect applicable law and third-party rights, and must not upload or generate content infringing copyright, trademarks or privacy. The service's own components remain the intellectual property of VideoFaceSwap and cannot be reused without prior written agreement. The pricing FAQ says most packages include a commercial-use licence for the outputs you generate, while leaving you to obtain every necessary permission. Uploaded media is processed temporarily and deleted, not kept as a publisher asset.
Reuse rights
The privacy policy lists what is collected: email, name, a phone number if you sign in by phone, the media you upload, and payment details handled by a third-party processor. Usage data is gathered automatically: IP address, browser type and version, device type and operating system, date and time of access, referring URL and device identifiers. The declared purposes are narrow: providing and maintaining the service, managing the account, contacting the user and handling their requests. Three uses are explicitly excluded: training AI models, advertising or profiling, and selling personal information to third parties. Sharing is said to be limited to cloud providers, payment processors and authorities where the law requires, all contractually bound. Cookies and analytics cover preferences, traffic analysis and fraud detection, with the analytics tools described as privacy-conscious. One internal contradiction deserves attention: the same policy describes ad network partners using cookies and pixels to serve targeted advertising, with an opt-out mechanism, immediately after ruling advertising and profiling out. The terms add that third-party technologies, including the ChatGPT API, are governed by their own conditions.
Data retention & training
Hosting summary
The privacy policy states that the services operate from Hong Kong and may involve cloud providers outside the user's country, so personal data may be transferred internationally. Beyond that, no hosting country and no region is ever named: the location of the servers is never specified. International transfers are said to be framed by Standard Contractual Clauses for the EEA and by data processing agreements with all third-party providers. Declared security measures are encryption of uploaded content in transit and at rest, access control with authentication for sensitive operations, and regular security audits and monitoring; the pricing FAQ adds isolated compute, audit logs and periodic reviews, and mentions enterprise options such as SSO/SAML, IP allow-lists and DPAs. No named subprocessor list is published and no certification such as SOC 2 or ISO 27001 is claimed. One network observation, outside anything the site declares: the resolved IP address 104.21.43.192 is a Cloudflare anycast node, which says nothing about where the data actually resides.
Things to keep in mind
Risks and trade-offs to weigh before adopting Deepfake Maker.
- The terms read as if they were written for another product: they name VideoFaceSwap Inc and refer to VideoFaceSwap throughout, while the site is called Deepfake Maker, a strong sign of documents reused from elsewhere
- The documents disagree with each other and carry three different vintages, terms updated 1 September 2024, privacy policy 1 July 2025, refund policy 1 January 2026, with auto-renewing monthly or annual subscriptions on one side and one-off credit purchases on the other
- Two different retention promises coexist: the home page states that uploaded media and generated content are never stored, while the privacy policy announces deletion within 24 hours
- The privacy policy excludes advertising and profiling, then describes ad network partners using cookies and pixels for targeted advertising, and the terms' disclaimer lists prohibitions modelled on mainland Chinese law, including constitutional principles, national security and heroes and martyrs, while the stated governing law is Hong Kong's
- Consent for the faces you use rests entirely on you, including in criminal terms, and producing a convincing swap in seconds lowers the threshold for casual misuse: a joke shared without thinking can be indistinguishable from an attempt to deceive
- The catalogue includes suggestive effect generators and an Image to Video NSFW use case promoted on the site, while the terms restrict the service to personal study and academic research and forbid sharing generated content without permission, which the entire product pitch contradicts
- Figures and identity are thin: a self-declared 4.9 rating over 89,868 reviews in the page's structured data and a claim of more than a million creative teams on the pricing page, no postal address and no legal name matching the brand, no price visible without JavaScript or an account, an Available Across Platforms block announcing Windows, Mac, iOS and Android without a single download link, and a sole contact email obfuscated by a Cloudflare token, invisible to a direct reading of the page
Setup & Integrations
Technical difficulty
Very low. Nothing is installed: everything runs in the browser, on desktop, tablet or mobile. The swap tool is embedded on the home page and usable without any prior navigation, and free use is claimed without registration. The workflow is three steps, and the product pages repeat that no editing skills are required. The only settings exposed are the photo, video and GIF tabs, the Single, Multiple and Batch sub-tabs, the Basic or HD model selector and the HD Enhance toggle. An account is needed only to buy credits, reach the history and unlock the advanced features.
Deployment
Integrations
Supported languages
Behind Deepfake Maker
Resources
All the official URLs gathered for verification and reference.
Frequently asked questions
What exactly is Deepfake Maker?
Do I need an account to use it for free?
Which file formats are accepted?
How long does one generation take?
Are my uploaded files kept?
Is my data used to train AI models?
How do the credits work?
Can I be refunded?
Can I use the results commercially?
Is it legal to use, and what are the rules?
Should you pick Deepfake Maker?
Deepfake Maker's strength is breadth and immediacy. The swap tool sits on the home page, works from any browser without installation, and reaches well beyond face swapping into image generation, video generation, restoration, upscaling, filters and effects. A permanent free tier exists, credits are advertised as never expiring, and purchases are described as one-off rather than as subscriptions. The reservations are almost all about transparency. No price is published: the pricing cards are filled in client-side, and nothing in the served page, the FAQ or the refund policy carries a figure, so the cost cannot be compared before creating an account. The publisher's identity stays blurred, an unnamed company based in Hong Kong in the privacy policy, an entity called VideoFaceSwap Inc in the terms, no postal address and no identified director. The three legal documents carry different dates and contradict each other on subscriptions against credits, on retention and on advertising. The privacy commitments themselves read seriously on paper, no training on user data, deletion within 24 hours, standard contractual clauses and processing agreements, but nothing is certified or audited. Finally, the legal and ethical risk inherent to deepfakes is transferred wholesale to the user: the FAQ makes consent your obligation, while the terms restrict the service to personal study and academic research even as the product is sold as a social sharing machine. The verdict follows that split. It is a reasonable candidate for social creation, memes and short-form work, and for creative prototyping where speed matters more than paperwork. It is a poor candidate for any procurement-led purchase, for journalistic or evidentiary work, or for any team that needs a named counterparty, a published price and an audited data trail.
- Choosing a selection results in a full page refresh.
- Opens in a new window.