Tether, the issuer of the top stablecoin “$USDT,” has officially unveiled QVAC MedPsy. QVAC MedPsy is an exclusive group of open-source medical AI models specified for deployment on laptops, smartphones, and the rest of the edge devices. As per Tether’s official press release, the development denotes a key step in advancing its wider AI strategy within the QVAC network. Additionally, MedPsy frameworks are focused on the provision of expert-scale healthcare reasoning without any reliance on costly enterprise hardware or cloud servers.
8 billion humans deserve an intelligence that doesn't blink when the signal dies. ?
— Tether (@tether) May 7, 2026
Introducing @QVAC Psy, our foundational models built on the mathematical stability of Psychohistory.
With QVAC MedPsy, our local-first medical health AI model, we’ve proven that superior… pic.twitter.com/6ECt7kvk6Q
Tether Launches QVAC MedPsy to Enable Decentralized Medical AI on Edge Devices
With the launch of QVAC MedPsy, Tether is providing sovereign and decentralized infrastructure for efficient healthcare reasoning with no dependence on capital-intensive enterprise-scale hardware or cloud servers. These lightweight models, compatible with edge devices, including laptops and smartphones, can outcompete the bigger medical AI mechanisms while also maintaining solid privacy protections via completely local inference.
The launch underscores the growing merger between decentralized technologies and healthcare, while AI institutions are competing to develop effective device-based intelligence solutions. The exclusively introduced QVAC MedPsy lineup takes into account a couple of text-only medical language frameworks, including MedPsy-4B and MedPsy-1.7B. The AI research group of Tether Data points out that these models offer effective “edge deployment,” permitting medical AI instruments to work on user hardware rather than centrally controlled cloud infrastructure.
The firm stressed that the frameworks are completely open-source under the Apache 2.0 license for educational and research use. A crucial feature of this move is the attention given to parameter efficiency. The firm claims that the lighter 1.7B model gained a 62.62 score on average across 7 medical standard suites.
Outcompeting Bigger Medical AI Frameworks With Comparatively Fewer Parameters
Hence, it outperformed MedGemma-1.5-4B of Google, irrespective of its comparatively less than 50% size. In the meantime, MedPsy-4B Framework also surpassed it on many healthcare evolutions while utilizing almost one-seventh of parameters. According to Tether , the new project also prioritizes privacy-centered healthcare deployment while strictly complying with GDPR, HIPAA, and other such regulations. Overall, with QVAC MedPsy, the platform is elevating its position within the AI infrastructure landscape, specifically in areas dealing with edge computing and decentralized intelligence.
