Existing HPC, AI, graph analytics, and blockchain compute solutions are no longer “good enough”.  

They are expensive, hard to program, power hungry, and only work for limited applications.

The Third Wave of Accelerated Computing is Here

Thunderbird for HPC, AI, Graph Analytics


Setting New standards of compute power, energy efficiency and versatility for a range of applications

InspireSemi is dedicated to delivering superior solutions for HPC, AI, graph analytics, blockchain and other compute-intensive applications. Led by an accomplished team with a proven track record, we are currently developing our next generation accelerator solution. It features a versatile sea-of-cores processor architecture with an established software ecosystem, and sets new standards of performance, ease of use, and energy efficiency.


Enabling tomorrow’s innovation today

Addressing the Need to Accelerate All HPC & AI Software

What customers always wanted…Not “yet another GPU”

InspireSemi’s powerful Thunderbird “supercomputer cluster-on-a-chip” accelerated computing architecture is based on an array of thousands of modern, efficient, and powerful 64-bit CPU cores, tightly integrated with high speed memory and networking.  It is an ideal and versatile platform for high-performance computing (HPC), AI/machine learning, graph analytics, and other compute-intensive applications.

Leveraging the modern, open standard RISC-V CPU ISA (instruction set architecture) provides access to an established, thriving ecosystem of software, libraries, and development tools. Our developer-friendly all-CPU programming model eliminates vendor lock-in and greatly simplifies software development, QA, and maintenance, since there is no need to support multiple and often proprietary software stacks (e.g.- x86, CUDA, ROCM, FPGAs, etc.). This also enables the combination of HPC & AI on one platform for deeper insights and bigger advancements.

Our highly optimized, high speed mesh network fabric provides extreme bandwidth and low latency communications between cores. Tightly integrated with the CPU cores and distributed memory, it removes crucial bottlenecks from applications that depend on close cooperation between many independent threads. Arrays of hundreds of up to 256 Thunderbird chips connect seamlessly via high-speed serial transceivers.

Compared to a modern datacenter GPU, Thunderbird provides all the raw computational power with far broader application to real world codes. Our advanced network fabric enables far better cooperation between cores. And our roots in extremely power-sensitive blockchain computing enable us to provide this performance with considerably higher energy efficiency than leading GPU’s.


Raw performance: 24 TFLOPS (FP64)
Energy efficiency: 50 GFLOPS/Watt (FP64)
Utilization: network reduces latency 20x
Deterministic + Predictable Performance
Value: exceptional Performance/$

  • Thousands of powerful 64-bit RISC-V superscalar CPU cores can leverage an established, thriving software ecosystem to support broader range of applications than GPUs


  • Developer-friendly CPU programming model greatly simplifies software & algorithm development, eliminates vendor lock-in, and provides maximum versatility for HPC, AI, Graph Analytics, and blockchain applications


  • Superscalar, vector, and tensor operations, including mixed precision floating point


  • Dedicated acceleration for several cryptography algorithms


  • Ample high-speed SRAM memory tightly integrated with processor cores


  • Extreme-bandwidth, low-latency on-chip mesh network fabric for both inter- and intra-chip connectivity, up to 256 Thunderbird chips


  • Deterministic + predictable performance addresses applications GPUs cannot, including real-time and real-safe computing


  • Support for large amounts of onboard LPDDR memory, NVMe storage, PCIe, and Gb ethernet

Download Executive Summary

Innovate with InspireSemi

Contact us to learn more about partnership opportunities