Avnet launches QCS6490 Vision-AI development kit

17th May 2024

Avnet Abacus

Harry Fowle

0 0

The QCS6490 Vision-AI Development Kit from Avnet enables engineering teams to rapidly prototype hardware, application software and AI enablement for multi-camera, high-performance, Edge AI-enabled custom embedded products.

Engineered in a compact form factor, the kit facilitates design with the recently announced energy-efficient MSC SM2S-QCS6490 SMARC compute module based on the Qualcomm QCS6490 processor.

The QCS6490 processor uses energy-efficient 6nm process technology, with 8 high-performance (A78 and A55) processor cores, a high throughput Spectra-570L triple ISP image processor, Qualcomm Adreno 633 VPU supporting 4K30/4K60 encode/decode, Adreno 643 GPU, 6th Gen Qualcomm AI engine (12 TOPS) and a versatile range of peripheral interface options.

“In close collaboration with Qualcomm Technologies, Inc., Avnet created the Vision-AI Development Kit to help developers quickly evaluate our production-ready SMARC compute module based on the QCS6490 processor.” said Jim Beneke, Vice President of Advanced Applications Group, Avnet. “The development kit is ideal for prototyping advanced, multi-camera vision-based systems and its dedicated AI engine helps enable edge-based generative AI using emerging LLM models that have been optimised for Qualcomm IoT processors.”

"Qualcomm Technologies is pleased to collaborate with Avnet on the launch of the Vision AI development kit. This kit enables developers to harness the power of the Qualcomm QCS6490 processor, with impressive compute and AI capabilities, to build cutting-edge Edge AI solutions,” said Sebastiano Di Filippo, Senior Director, Business Development, Qualcomm Europe, Inc. "We believe that this technology will unlock significant opportunities for the IoT industry, and we are excited to see what developers will create with this kit."

The vision-AI development kit includes:

Support for up to four MIPI CSI cameras
Support for concurrent Mini DisplayPort and MIPI DSI displays
An audio subsystem with two onboard microphones, audio codec, digital and analog I/O
1-Gbit Ethernet plus optional Wi-Fi 6E networking
Two USB3.1 and three USB2.0 interfaces
Two CAN interfaces
Pi HAT-compatible 40-pin expansion
Two M.2 expansion interfaces for storage, acceleration and wireless options