Ai acceleration co-processor IP

Instruction set: RISC-V RV64GC/RV 64GCV;
Multi-core: Isomorphic multi-core with 1 to 4 optional clusters. Each cluster can have 1 to 4 optional cores;

Fully programmable to efficiently execute Neural Networks, feature extraction, signal processing, audio and control code
Scalable performance by design to meet wide range of use cases with MAC configurations with up to 64 int8 (native 128 of 4x8) MACs per cycle
Future proof architecture that supports the most advanced ML data types and operators

Fully configurable comprehensive IP platform for 5G NR and multi-mode RAT addressing both Basestation and Radio Open RAN use cases

Fully configurable IP platform for 5G NR and multi-mode RAT.
Dedicated and optimized Max configuration for eMBB and Lite configuration for emerging cellular IoT modems.

Cycle and energy efficient for Bluetooth and Bluetooth Low Energy (BLE) codecs for speech and music
Efficient neural network acceleration ISA and architecture support

Complete HDR ISP video processing framework for multi-channel vision and AI systems
Demonstrates logicBRICKS HDR ISP pipeline for parallel processing of three UHD automotive video cameras
Fully compatible with Xylon logiIVID-ACAP-ISP HDR ISP Evaluation Kit for Versal ACAP based on AMD-Xilinx ACAP Versal VCK190 Evaluation Kit
Runs on Linux OS, includes logicBRICKS software drivers and demo applications made with AMD-Xilinx Vitis™ Unified Software Platform 2021.2

AI Acceleration Co-Processor IP