Firefly's latest NVIDIA Jetson Orin motherboard is equipped with the Jetson Orin series core modules, with a computing power of up to 157 TOPS. It supports the private deployment of large language models and provides strong computing power support for the development and deployment of edge AI applications.
High-Performance Edge Computing System On Chip
Equipped with the NVIDIA Jetson Orin NX edge computing module (16GB version), it is equipped with an eight-core ARM CPU and a 1024-core NVIDIA Ampere architecture GPU with 32 Tensor Cores. It can run multiple concurrent AI application pipelines and provide powerful AI performance.
Up To 157 TOPS Computing Power
AIO-OrinNX and AIO-OrinNano have computing powers of 157 TOPS and 67 TOPS respectively, capable of running current mainstream AI models, realizing larger and more complex deep neural networks, and have functions such as object recognition, target detection and tracking, speech recognition, and other visual development, adapting to a variety of AI application scenarios.
Private Deployment Of Large Language Models
It supports the private deployment of modern mainstream AI, such as large language models such as LLaMa2, ChatGLM, Qwen, large visual models such as ViT, Grounding DINO, SAM, and supports the Stable Diffusion V1.5 image generation model in the AIGC field.
Supports Up To 8K Video Decoding
AIO-OrinNX supports 18-channel 1080p30 H.265 video decoding and can support up to 1x8K30 (H.265), 2x4K60 (H.265), 4x4K30 (H.265), 9x1080p60 (H.265), and 18x1080p30 (H.265) video decoding to meet the needs of various AI application scenarios.
Generative AI At The Edge
The NVIDIA Jetson Orin series provides powerful AI computing capabilities, large unified memory, and a comprehensive software stack to drive the latest generative AI applications with ultra-high energy efficiency. It enables fast inference for any generative AI model powered by the Transformer architecture, achieving excellent edge performance on MLPerf.
AI Stack And Ecosystem
Democratize edge AI and robotics development with a comprehensive AI software stack and ecosystem powered by Generative AI at the Edge and the NVIDIA Metropolis and Isaac platforms. With NVIDIA JetPack, Isaac ROS, and reference AI workflows, you can integrate advanced technologies into your products without relying on in-house AI resources.
Extensive Application Scenarios
It is widely used in products and fields such as enterprise security, traffic management, public safety, environmental protection, smart cities, and home security.
