The Semiconductor IP Marketplace

Try an example: PCIe 7.0
☰ Semiconductor IP

The Pulse

The Semiconductor IP Marketplace that puts you first

Semi IP Hub's mission is to provide you with a platform where you can find Silicon IP cores for your next project without being harassed by dozens of sellers.

Here, your contact details are not shared with third parties unless you request to be contacted by a supplier.

Spotlight

  • High Bandwidth Memory 3 (HBM3/3E) IP optimized for Samsung SF4X
    • The HBM3 IP consists of a PHY and memory controller optimized for Samsung SF4X process to support the HBM3 memory standard (JESD238A) operating at up to 9.6 Gbps/pin.
    • The HBM3 IP is designed for high memory throughput and low latency applications while minimizing area and power consumption.
    Block Diagram -- High Bandwidth Memory 3 (HBM3/3E) IP optimized for Samsung SF4X
  • Crystal Oscillators
    • The crystal oscillator macros are available in a wide range of industry-standard quartz crystals and MEMS resonators operating in the fundamental mode in the 32 kHz to 80 MHz range.
    • These oscillators, which are both power and area efficient, have a programmable transconductance to allow users to find the optimal balance between jitter and power consumption.
    Block Diagram -- Crystal Oscillators
  • SENT/SAE J2716 Receiver
    • The CSENT-RX core implements a receiver for the Single Edge Nibble Transmission (SENT) protocol.
    • It complies with the SAE J2716 standard and supports both synchronous and asynchronous sensors.
    • It can be used for receiving data from one or multiple sensors using a single SENT line.
    Block Diagram -- SENT/SAE J2716 Receiver
  • LLM Accelerator IP for Multimodal, Agentic Intelligence
    • HyperThought is a cutting-edge LLM accelerator IP designed to revolutionize AI applications.
    • Built for the demands of multimodal and agentic intelligence, HyperThought delivers unparalleled performance, efficiency, and security.
    Block Diagram -- LLM Accelerator IP for Multimodal, Agentic Intelligence
  • ORAN IP core
    • ORAN IP core is a highly scalable and silicon agnostic implementation of the interoperable O-RAN WG4 defined 7.2x interface for deployment in O-DU and O-RU products, targeting any ASIC, FPGA or ASSP technologies.
    • The ORAN over eCPRI implementation builds on long-time experience designing CPRI and Radio-Over-Ethernet solutions for fronthaul and delivers a flexible engine that is prepared for tight integration with software applications.
    Block Diagram -- ORAN IP core
  • E-Series GPU IP
    • E-Series GPU IP delivers fast and flexible parallel compute that scales from wearables to the cloud.
    • E-Series represents a new era of GPU IP with the introduction of a lot of dense, deeply integrated acceleration for power-efficient AI operations – up to 4x more than Imagination D-Series GPU IP.
    Block Diagram -- E-Series GPU IP

UCIe Controller IP View All

  • UCIe Die-to-Die Chiplet Controller
    • High configurability and customizability
    • Defines packets to communicate with a link partner using different AXI parameters
    • Supports raw streaming modes
    • Provides various Flit formats in UCIe v1.1 (filt format 2: 68B flit format, flit format 3/4: standard 256B flit format, and flit format 5/6: latency optimized 256B flit format)
    Block Diagram -- UCIe Die-to-Die Chiplet Controller
  • UCIe PHY & D2D Adapter
    • 32Gbps UCIe-Advanced (UCIe-A) & Standard (UCIe-S)
    • UCIe v1.1 specification
    Block Diagram -- UCIe PHY & D2D Adapter
  • 2-16Gbps Multi-Protocol IO Supporting BOW, OHBI and UCIe
    • High Bandwidth Density and Data Rates
    • Package Configurability
    • Energy Efficiency
    • Fully Integrated Solution
    Block Diagram -- 2-16Gbps Multi-Protocol IO Supporting BOW, OHBI and UCIe

PCIe 7.0 IP View All

  • PCIe 7.0 Controller with AXI
    • Optimized for high-bandwidth efficiency at data rates up to 128 GT/s
    • Separate native TX/RX data path separating posted/Non posted/completion traffic
    • Handles up to 4 TLPs per cycle
    • Advanced PIPE modes and port bifurcation
    Block Diagram -- PCIe 7.0 Controller with AXI
  • PCIe 7.0 PHY IP
    • Physical Coding Sublayer (PCS) block with PIPE interface
    • Supports PCIe 7.0, encoding, backchannel initialization
    • Lane margining at the receiver
  • PCIe 7.0 PHY in TSMC (N5, N3P)
    • Physical Coding Sublayer (PCS) block with PIPE interface
    • Supports PCIe 7.0, encoding, backchannel initialization
    • Lane margining at the receiver
    • Spread-spectrum clocking (SSC)

AI IP View All

  • RISC-V CPU IP
    • RISC-V RVA23 Compliant 
    • >18 SPECint2006/GHz
    • 8-wide decode unit
    • Advanced branch predictor
    Block Diagram -- RISC-V CPU IP
  • NPU IP for Embedded ML
    • 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
    Block Diagram -- NPU IP for Embedded ML
  • Future-proof IP for training and inference with leading performance per watt and per dollar
    • RISC-V-based AI IP development for enhanced training and inference.
    • Silicon-proven solutions tailored for AI workload optimization.
    • Energy-efficient performance with industry-leading Perf/W.
    Block Diagram -- Future-proof IP for training and inference with leading performance per watt and per dollar

MIPI IP View All

RISC-V IP View All

×
Semiconductor IP