If you need AI device we are the best place to get it.
Find the right option for your use case.
ESP32 | AIoT Solution
Ideal for devices requiring minimal local processing
The go-to solution for cloud-based AI services
Ultra-low power consumption for extended battery life
Perfect for smart home devices, wearables, and sensor networks use cases
Compute Module 4 | Model < 1B
Optimized for small size language models
Up to 10 tokens/sec for running TinyLlama real time
Good balance of power consumption and compute power
Suitable for smart speakers, industrial IoT, and entry-level AI assistants
Compute Module 5 | Model < 7B
Good performance computing for medium-sized language models
> 8 tokens/sec for running Phi-3B real time
Ideal for complex LLM tasks and vision model applications
Perfect for advanced robotics, autonomous systems, and llm-powered kiosks
Jetson Nano | Model >7B
GPU-accelerated computing for state of the art AI models
Up to 15 tokens/sec for running Llama3-8B real time
Suitable for running multiple AI models simultaneously
Suitable for advanced security systems, office assistant and offline agent system