Tiiny AI provides the Pocket Lab, a 200 g plug‑and‑play USB‑C edge compute module with an integrated GPU/TPU accelerator that runs full‑stack AI inference locally without subscription or token fees. The device includes a Python/ONNX SDK, secure boot, encrypted storage, and OTA model and firmware updates, targeting developers and small‑to‑mid‑size enterprises that require low‑latency, on‑device AI.
Funding
Funding not disclosed
Founders
Product
Problem
Many edge applications require on-device artificial intelligence but are constrained by reliance on cloud services, subscription fees, or token‑based pricing models, which increase latency, cost, and data‑privacy risks. Existing hardware solutions are often bulky, expensive, or locked into proprietary ecosystems, limiting accessibility for developers and small enterprises.
Solution
Tiiny AI delivers the Pocket Lab, a compact, plug‑and‑play hardware accelerator that runs full‑stack AI workloads locally without any recurring subscription or token costs. Users can download pretrained models or deploy custom agents directly onto the device, leveraging an integrated GPU/TPU‑style inference engine optimized for low‑power operation. All inference runs offline, ensuring sub‑millisecond response times and preserving data sovereignty. The device ships with a secure bootloader, encrypted storage, and a lightweight SDK that abstracts hardware details, enabling rapid integration into existing software stacks. Firmware updates are delivered over the air, keeping the platform current without additional fees. By offering a one‑time purchase model, Tiiny AI eliminates ongoing operational expenses and simplifies budgeting for edge AI deployments.
Target Audience
The primary customers are independent developers, IoT startups, and small‑to‑mid‑size enterprises that need reliable, low‑latency AI at the edge without recurring licensing costs.
Features
- Portable edge compute module (≈200 g) with integrated GPU/TPU accelerator supporting FP16/INT8 inference
- On‑device model repository with seamless OTA model download and version control
- Full Python/ONNX runtime SDK for rapid deployment of custom agents and pipelines
- Secure boot and encrypted storage to protect intellectual property and user data
- Power‑efficient design with configurable performance modes for battery‑operated use cases
- Plug‑and‑play USB‑C interface with driver‑less operation on Windows, macOS, and Linux
- Over‑the‑air firmware updates and diagnostics via a web‑based console