DEEPX builds physical AI semiconductor solutions that deliver GPU‑level accuracy with ultra‑low power consumption and heat, enabling high‑performance inference on edge and battery‑powered devices. Their DX‑M1 accelerator provides 240% performance over a 40 W GPU at just 5 W, while the upcoming DX‑M2 targets generative AI workloads, and IQ8 quantization offers INT8 speed with FP32 precision, reducing total cost of ownership by up to 94%.
Funding
$103.6M raised to dateRaised to date based on public sources. This may differ from the amount the company actually raised and is based only on what is publicly available on the internet.
Founders
Product
Problem
Deploying advanced AI models on edge devices is limited by high power consumption, heat generation, and the total cost of ownership of traditional GPUs, making on-device inference expensive and inefficient.
Solution
DEEPX provides purpose-built physical AI semiconductor solutions that deliver GPU‑level accuracy with dramatically lower power draw and heat output. Their DX‑M1 chip achieves 240% performance over a 40 W GPU while consuming only 5 W, and the upcoming DX‑M2 extends this efficiency to generative AI workloads on battery‑powered devices. DEEPX’s IQ8 intelligent quantization enables INT8‑speed inference with FP32‑grade precision, reducing energy use and emissions by up to 90%. The company offers a portfolio of AI chips, modules, and complete systems that integrate easily into edge, smart mobility, factory, and city applications, delivering high‑performance inference with a total cost of ownership up to 94% lower than conventional GPUs.
Target Audience
Primary customers are OEMs and system integrators building edge AI solutions for smart mobility, industrial automation, smart city infrastructure, and other battery‑operated or low‑power devices.
Features
- DX‑M1 AI accelerator: 5 W power envelope, 20× efficiency improvement, 240% performance over a 40 W GPGPU
- IQ8 quantization: INT8 execution speed with FP32 accuracy for maximum model fidelity
- DX‑M2 (in development): optimized for on‑device generative AI and large language models at battery‑level power
- Compact AI modules and full AI systems designed for minimal board space and easy integration into IoT and edge products
- Ultra‑low heat generation and power consumption reduce cooling requirements and operational costs
- Proven reductions in energy use and carbon emissions equivalent to planting thousands of trees