PRODUCTS
Infertron-E/P :
Extremely efficient On-sensor/On-things AI
Infertron-E/P is a cutting-edge hardware solution designed for extremely efficient AI inference, offering optimized performance and energy consumption.
Its compact and energy-efficient architecture makes it ideal for edge AI, IoT devices, and applications requiring embedded intelligence in minimal spaces.
| Model | Infertron-E | Infertron-P |
|---|---|---|
| Phase in roadmap | Phase I | |
| Application | On-sensor / On-things | |
| Characteristics | Minimum Cost & Energy, > 60FPS | Low Cost & Energy |
Performance
Target application
On-sensor AI
On-things
Initial target application : mobile & edge H/W
Due to the low power, low cost, small area, APU can replace GPU/TPU and create new mobile H/W markets such as image processing, voice processing, security & surveillance, consumer commercial drone, objective recognition, autonomous cars.
Image Processing
Voice Processing
Commercial Drones
Autonomous Car
Pattern Recognition
Security & Surveillance
Robot
Infertron-S/U :
High performance AI solution with low cost and energy
Infertron-S/U deliver exceptional AI capabilities without compromising on energy consumption or cost.
The optiomal solutions for low-cost, high-efficiency inference of large-scale detection models and LLMs (Large Language Models), offering un-paralleled performance for modern AI applications.
| Model | Infertron-S | Infertron-U |
|---|---|---|
| Phase in roadmap | Phase I & II | Phase II & IIII |
| Application | On-device | Server / HP-Edge |
| Characteristics | High performance / Low Cost / Low energy | Ultra performance / Low Cost / Low energy |
Target application
High performance on-device AI detec-tion model
LLM-based application for server pro-viders
etc
