Access breakthrough compute performance InspireSemi™ delivers blockchain, HPC and AI accelerators that beat the benchmarks for speed, ease of programming, energy efficiency, and bottom-line value. This gives innovators in compute-intensive industries the performance, efficiency, and time to market advantages they need to drive exceptional outcomes. Learn More

Existing HPC, AI, and blockchain compute solutions are no longer “good enough”.
They are painfully slow, hard to program, expensive, and power hungry.

Blockchain operations
Maximize the profitability of your blockchain operations, including crypto currency and mining, by tapping into the superior performance and energy efficiency of our next generation accelerator chip. InspireSemi has built on the success of our Gen1 ASIC with a fully programmable Gen2 chip that includes dedicated acceleration for key blockchain algorithms. Blistering speed, energy efficiency, a standard CPU programming model, and game-changing affordability make it an ideal choice for blockchain applications.
High Performance Computing
Engineering, Scientific and Life Sciences applications achieve new standards of performance, scale, and efficiency, including ones that did not benefit from GPU accelerators. Our next generation accelerator leverages thousands of 64 bit RISC-V processor cores interconnected with an extreme-bandwidth, low-latency on-chip mesh network fabric. Key HPC compilers, libraries and tools are supported via the rich RISC-V open source ecosystem. The developer-friendly CPU programming model greatly simplifies software development and eliminates vendor lock-in (e.g.- CUDA, ROCM, FPGAs, etc.). This also simplifies and enables the convergence of HPC+AI for even bigger advancements.
AI/ML
developers
Unleash your AI applications from data bottlenecks and undue expense to radically accelerate innovation, insights, and achieve greater market impact. Our next generation accelerator leverages thousands of 64 bit RISC-V processor cores interconnected with an extreme-bandwidth, low-latency on-chip mesh network fabric that enables very large memory models. All major AI frameworks and tools are supported via the rich RISC-V open source ecosystem. The developer-friendly CPU programming model greatly simplifies software & algorithm development and eliminates vendor lock-in (e.g.- CUDA, ROCM, proprietary AI chips, etc.). And it easily enables the convergence of AI+HPC for greater productivity and deeper insights.

InspireSemi

MISSION

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

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

SOLUTIONS

Enabling tomorrow’s
innovation
today

InspireSemi’s powerful “best of the best” processor architecture is based on an array of thousands of simple but powerful CPU cores, tightly integrated with high speed memory and networking.  It is an ideal platform for high-performance computing (HPC), AI/machine learning, and ultra-efficient blockchain applications.

The RISC-V instruction set provides access to an established, thriving ecosystem of open-source software 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 convergence of HPC+AI (aka computational steering) for deeper insights and bigger advancements.

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

Compared to a modern datacenter GPU, our architecture provides all the raw computational power with far greater flexibility and agility. Our advanced network fabric enables far better cooperation between cores. And our roots in extremely power- and cost-sensitive blockchain computing enable us to provide this performance with considerably higher energy efficiency than leading GPU’s.

10X throughput
30-60% power reduction
Existing software ecosystem
70-90% below competitors’ pricing

Thousands of powerful 64-bit RISC-V CPU cores & established, thriving software ecosystem

Ample high-speed SRAM memory tightly integrated with processor cores

Superscalar, vector, and tensor operations

Dedicated acceleration for several blockchain algorithms

Extreme-bandwidth, low-latency on-chip mesh network fabric for both inter- and intra-chip connectivity

Support for seamless multi-chip arrays approaching one million cores via high-speed SerDes links

Support for large amounts of onboard DDR DRAM, SSD storage, PCIe, and 40G ethernet

Developer-friendly CPU programming model greatly simplifies software & algorithm development, eliminates vendor lock-in, and enables the convergence of HPC+AI

.

InspireSemi

Download Executive Summary


    Innovate with InspireSemi

    Contact us to learn more about partnership opportunities.