Tern provides a ternary‑logic computing platform that combines the Talc CPU, the Tidal SDK, and the Tree hardware system to deliver up to 25× higher storage density and roughly 50 % lower active power consumption for AI, robotics and cloud workloads. The unified Cogence stack enables native ternary code development, emulation, AI‑assisted optimization, and built‑in post‑quantum cryptography, reducing latency and energy use while increasing data‑processing efficiency. The solution is offered via subscription and optional hardware leasing for data‑center operators, research labs, robotics manufacturers and cloud providers.
Funding
Funding not disclosed
Founders
Product
Problem
Current binary computing architectures create performance bottlenecks that increase latency, energy consumption, and storage overhead for workloads such as AI model training, robotics control, and large‑scale data‑center processing. These constraints limit the ability of enterprises to scale compute‑intensive applications cost‑effectively.
Solution
Tern replaces binary logic with native ternary computing through its Cogent platform. The Talc CPU implements the world’s first ternary‑logic semiconductor, delivering up to 25× higher storage density and cutting active power use by roughly 50 % while improving data efficiency by 37 %. The Tidal SDK provides a full software stack that enables developers to write, compile, and debug ternary code, with built‑in emulation (Hardsim) and AI‑assisted guidance (Cyrus). Tree, the flagship hardware system, integrates Talc processors and the Cogence runtime to offer a turnkey, low‑latency compute node for AI, robotics, and cloud workloads. Together, these components reduce compute overhead, accelerate inference and training cycles, and support post‑quantum cryptographic security, allowing organizations to scale heavy workloads with substantially lower energy footprints.
Target Audience
Primary customers are AI research labs, robotics manufacturers, cloud service providers, and data‑center operators that require high‑performance, energy‑efficient compute, as well as software developers building next‑generation applications on ternary architecture.
Features
- Talc ternary‑logic semiconductor: 25× storage density increase, 50 % active power reduction, 33 % passive power reduction, 37 % data‑processing efficiency gain.
- Tidal SDK: native ternary programming language, API‑first design, Hardsim emulator, AI assistant (Cyrus) for code guidance, and Explorer platform for project sharing.
- Tree hardware platform: integrated Talc CPU, Cogence runtime, and high‑throughput I/O for low‑latency AI and robotics workloads.
- Unified Cogence software stack: supports AI acceleration, robotics automation, cloud‑edge distribution, and post‑quantum cryptography across all products.
- Scalable licensing model: subscription‑based access to Tidal Pro features and optional hardware leasing for enterprise deployments.
- End‑to‑end security: built‑in PQC primitives and hardware isolation to protect data in transit and at rest.