NETWEB Technologies India launched its ‘Make in India’ AI Supercomputer, the Tyrone Camarero GB200 System, and the Tyrone Camarero Spark, one of the world’s smallest AI supercomputers, in collaboration with NVIDIA to boost AI development in India.
NETWEB Technologies India Ltd, a leading Indian OEM in high end computing, on Wednesday powered a new era of computing in India by introducing one of the world’s most powerful AI infrastructure solutions — a ‘Make in India’ AI Supercomputer Tyrone Camarero GB200 System — and the petascale personal compute system Tyrone Camarero Spark.
Tyrone Camarero Spark: A Desktop AI Supercomputer
According to a company statement, Netweb today announces a new class of AI Computing for India with the launch of Tyrone Camarero Spark, which is one of the world’s smallest AI supercomputers, delivering NVIDIA’s AI stack in a compact desktop form factor. It combines NVIDIA Blackwell GPUs, NVIDIA Grace CPUs, NVIDIA Networking, NVIDIA CUDA-X libraries and NVIDIA AI software stack, accelerating agentic and physical AI development to address the requirements of millions of AI developers in India. The system packs a petaflop of AI performance and 128GB of unified memory into a compact desktop form factor-based system, helping a huge customer base of AI developers in India to run inference on AI models with up to 200 billion parameters and locally fine-tune models of up to 70 billion parameters. It also allows developers to build AI agents and operate advanced software stacks entirely on-premises.
“The Tyrone Camarero Spark based on NVIDIA DGX Spark and powered by the NVIDIA GB10 Grace Blackwell Superchip, manufactured in India by Netweb and ready to serve the AI developer market of the world, is a proud moment for the ‘Make in India’ mission. The Tyrone Camarero Spark provides AI developers t 1 Peta Flops of (FP4) performance out of the box in the tiniest form factor of 5.9 in x 5.9 in x 2in inches and with system power supply of just 240 Watts –whereas in 2016 a comparable performance of 170 Tera Flops (FP16) with same memory size of 128GB could be delivered with System Power requirement of 3200 Watts. This milestone revolutionizes AI compute in India by introducing new ways to conduct AI research and development. With this move we have created a whole new customer segment in collaboration with NVIDIA.” said Sanjay Lodha, Managing Director and CEO of Netweb Technologies India Ltd.
“Increasing demand for generative AI and sovereign compute requires a new class of powerful, efficient infrastructure to accelerate innovation,” said Vishal Dhupar, Managing Director, South Asia, NVIDIA. “Netweb’s Tyrone Camarero Spark systems demonstrate how NVIDIA and India’s technology leaders are working together to deliver AI computing directly to the nation’s developers and enterprises to speed new discoveries.”
Tyrone Camarero Spark will provide AI developers an access to preinstalled NVIDIA Nemotron open models, libraries, NVIDIA NIM microservices with enabled workflows to help create vision search and summarisation agents, refining image generation, building an AI chatbot and much more. Shipping has already started and some of the units of this system are supplied to select AI user organisations in India.
‘Make in India’ AI Supercomputing: Tyrone Camarero GB200
Netweb is also launching Tyrone AI Supercomputing systems based on NVIDIA Grace Blackwell platforms designed and manufactured in India bolstering the remarkably successful Make in India mission. This product and architecture, built on the NVIDIA GB200 NVL4 platform, delivers revolutionary performance through four NVIDIA Blackwell GPUs connected with an NVLink bridge and two NVIDIA Grace CPUs connected over NVLink-C2C. Compatible with liquid-cooled, NVIDIA MGX™ modular servers, it provides up to 2x performance for scientific computing, AI model training, and inference over the prior generation.
Advanced Features and Capabilities
Netweb’s Tyrone Camarero GB200 AI System introduces many revolutionary technologies, which together enable AI training and real-time LLM inference for models scaling up to 10 trillion parameters. These include NVIDIA GB200 NVL4 – based on one of the world’s most powerful AI chips, second-generation Transformer Engine, fifth-generation NVlink, RAS Engine, secure AI with confidential computing and a Decompression Engine.
Deep Collaboration and Rack-Scale Integration
“Our deep collaboration with NVIDIA for the last 15 years has been instrumental in advancing AI infrastructure in India with the nation’s own sovereign-built, latest-generation NVIDIA systems. Make in India Tyrone GB200 systems, coupled with complete sovereign cloud stack and utilities provide a multi-node, liquid-cooled, rack-scale system design for the most compute-intensive workloads,” said Sanjay Lodha, Managing Director & CEO of Netweb Technologies India Ltd.
“This design will integrate 20 Tyrone GB200 MGX systems within a single liquid-cooled rack, comprising 40 NVIDIA Grace CPUs and 80 NVIDIA Blackwell GPUs, along with NVIDIA BlueField-3 DPUs, high-performance storage, and comprehensive end-to-end security,” Lodha added.
Netweb will display a complete range of systems of the Tyrone AI product range including the Tyrone NVIDIA MGX-based GB200 liquid cooled solution and Tyrone Camarero Spark which will reflect Netweb’s Make in India strength, for AI edge to AI personal computing to the most advanced AI data centre computing workloads, in New Delhi between 16th-20th February 2026. (ANI)
(Except for the headline, this story has not been edited by Asianet Newsable English staff and is published from a syndicated feed.)