Swallow Incubate opens liveness detection API to developers worldwide

Swallow Incubate has launched its liveness detection REST API globally with a free tier, letting developers check whether a face image shows a live person or a spoof attempt in a single call. The cloud service is aimed at face recognition and eKYC systems, with self-serve signup, Tokyo hosting, and plans starting at $180 a month.

Why it matters: - Face recognition and online identity checks are vulnerable to simple photo-based spoofing attacks. - Swallow Incubate is now offering the same anti-spoofing technology used in production in Japan as a cloud API for developers worldwide. - The launch lowers the barrier to adding liveness checks because teams do not need to install a client library or manage per-OS binaries.

What happened: - Swallow Incubate Co., Ltd. launched its Liveness Detection API with the latest generation of its anti-spoofing model. - The service is open to developers worldwide and includes a free tier. - The API takes a single still image and returns a verdict on whether the face belongs to a live person or a presentation attack. - The release date is October 15, 2026.

The details: - Developers send a Base64-encoded JPEG or PNG image up to 10 MB to POST /v1/check/liveness. - The JSON response includes a real-or-spoof verdict, a spoofing likelihood score from 0.0 to 1.0, face brightness, and detected face position. - The service is operated with a target response time of 500 ms or less per request. - Swallow Incubate says the technology is already used in Japan in authentication solutions for major medical institutions, access control terminals, and time and attendance terminals. - The Free plan includes 50 calls per month for evaluation. - Paid plans start at $180 per month for 1,000 calls. - Only successful responses count toward usage. - Signup is self-service, with no sales call, quote request, or contract negotiation required. - Developers can send their first request the same day. - Plans, pricing, and sign-up are available here. - API reference, quickstart, and sample code are available here. - A demo video is available here.

Between the lines: - Swallow Incubate is positioning the product for quick adoption by teams that want liveness checks without infrastructure overhead. - The company is emphasizing generalization, or performance in unfamiliar capture environments and against new attack patterns, rather than only benchmark results. - An internal evaluation using approximately one million frames of the company's own test data found 99.9% detection accuracy at the standard decision threshold. - The company notes that the accuracy figure comes from internal testing and is not certified by a third-party accreditation body. - The service is hosted in Tokyo, Japan, and submitted images are held in memory only during processing. - Swallow Incubate says images are not persistently stored. - The service is designed with the company's obligations as a data processor under GDPR in mind. - A Data Processing Agreement is available on request. - The company is working toward ISO/IEC 27001 certification. - The service is operated with a target monthly uptime of 99.5%. - Swallow Incubate was founded in 2013 and is headquartered in Tsukuba, Japan. - The company holds patents jointly filed with Panasonic Corporation in Japan and overseas. - Its other products include blink detection, gaze detection, and iris recognition.

What's next: - Developers can start testing the API immediately through the free tier. - Swallow Incubate is continuing work toward ISO/IEC 27001 certification. - The company says release details are current as of publication and may change without notice. - More information is available on Swallow Incubate's social account.

The bottom line: - Swallow Incubate is turning a production anti-spoofing model into a self-serve cloud API, aiming to make liveness detection easier to adopt for authentication and identity workflows.