Next Silicon
Rethinking compute architectures for future processing needs with novel VLSI designs.
About
NextSilicon develops the Maverick-2, a production dataflow accelerator powered by their Intelligent Compute Architecture™. This hardware adapts in real-time to software, delivering breakthrough performance for HPC, AI, and vector database applications without code rewrites or vendor lock-in. They serve developers in HPC and AI by eliminating the need for domain-specific languages and extensive porting efforts.
Key facts
At-a-glance profile data
- Website
- nextsilicon.com
- Headquarters
- Tel Aviv, Israel
- Founded
- 2017
- Employees
- 370
- Total funding
- $235M (Series C)
- Business model
- B2B
- Tags
- ComputingVLSISemiconductorAI ChipAI Hardware
Funding rounds
Tech stack
Detected on nextsilicon.com, last scanned 22 Sept 2026. Run your own scan
- Frontend
- Angular
- Analytics
- Google Analytics 4, Google Tag Manager, Segment
- CMS / Builders
- Ghost, GitBook
- CDN
- Fastly
- Email provider
- Google Workspace
Frequently asked
How much funding has Next Silicon raised?
Next Silicon has raised a total of $235M in funding. The most recent round on record is Series C.
How many employees does Next Silicon have?
Next Silicon has approximately 370 people on record.
NVIDIA H100 Tensor Core GPU
NextSilicon's Maverick-2 accelerator directly competes with NVIDIA's H100 in providing high-performance computing and AI acceleration, aiming to offer superior performance and efficiency for similar workloads.
AMD Instinct MI300 Series Accelerators
NextSilicon's Maverick-2 competes with AMD Instinct MI300 series accelerators by offering an alternative hardware solution for HPC and AI workloads, with a focus on real-time adaptability and efficiency.
Intel Gaudi AI Accelerators
NextSilicon's Maverick-2 and Intel Gaudi AI Accelerators both target high-performance AI training and inference, with NextSilicon emphasizing its adaptive architecture as a differentiator.
Cerebras CS-3
NextSilicon's Maverick-2 and Cerebras CS-3 systems both aim to accelerate large-scale AI and HPC workloads, with Cerebras utilizing wafer-scale integration while NextSilicon focuses on a dataflow architecture.
Graphcore IPU
NextSilicon's Maverick-2 and Graphcore's IPU (Intelligence Processing Unit) both offer novel architectures designed for machine learning acceleration, providing alternatives to traditional CPU/GPU approaches.