The European Commission included Tachyum in an Important Project of Common European Interest (IPCEI), which fully recognizes the uniqueness of the Tachyum technology and its importance for the European Union.

Tachyum is among the 68 highly ambitious IPCEI projects, that aim at developing unique technologies that go beyond what the market currently offers and will allow major improvements.

The Commission has accepted the funding gap of 26.4 million EUR for Tachyum to accelerate the delivery of the Prodigy 2 Universal Processor for HPC/AI (High Performance Computing/ Artificial Intelligence) in a 3-nanometer fabrication process, which will enable exa-scale HPC and zetta-scale AI for Europe. This project addresses a fundamental tenet of the IPCEI multi-national program – the need for a competitive EU processor.

The European Commission has approved, under EU State aid rules, IPCEI to support research, innovation, and the first industrial deployment of microelectronics and communication technologies across the value chain and announced its participants. Together, the Commission approved up to 8.1 billion EUR support by 14 member states for the IPCEI in microelectronics and communication technologies (IPCEI ME/CT). Strengthening Europe’s capacity to develop the next-generation of processors and semiconductors is key to ensuring Europe’s technology sovereignty and competitiveness, as well as to addressing key environmental and societal challenges.

“We are pleased to be among the highly ambitious IPCEI projects chosen to help propel the EU beyond what the market currently offers and enable major improvements in the development of sovereign technologies.”

Dr. Radoslav Danilák

Artificial intelligence is not the future, but the present. After the boom of large language models (LLM), cognitive modeling could be successful, which would, for example, move autonomous cars closer to human-like behavior. However, such models require 10x more computing power than is available today. The growing demand for compute-intensive applications requires a massive expansion of data center capacity, which will require a massive increase in electricity consumption. We must find another way than building new power plants required to supply electricity to data centers in the coming years. Our concept, where AI and classic processing are unified in one chip might avert future energy use growth connected to data centers.

Recently, we released the second edition of our trends white paper on Tachyum Prodigy’s leading position within the AI Industry featuring updates such as the implementation of improved 8-bit floating point (FP8) quantization aware training and post-training quantization achieving 32-bit floating point (FP32) accuracy.

FP8 is essential for being able to do more with less. It achieves much higher performance at much lower power consumption and chip area than legacy technology like BFLOAT16. FP8 not only reduces the cost of computation but also memory requirements for large and rapidly growing AI models.

Visual models, large language models and generative models are increasingly being included in a number of software applications, making AI an essential part of data processing and requiring tighter and lower latency integration into software. With FP8 capable of performing mainstream AI functions, our Prodigy chips are poised to help accelerate the rapid evolution of AI hardware technology.

Prodigy is the most effective compared to other chips in video analysis. They are used, for example, in autonomous unmanned aircraft systems that need to analyze images from cameras at a speed of 5,000 kilometers per hour.

“We did not target the AI market, it just so happens that our chip is much more powerful in this segment than others. Our experimental results show that FP8 enables faster training and reduced power consumption without any degradation in accuracy for a range of deep learning models.”

Dr. Radoslav Danilák

Robert Cisar, Tachyum’s Business Development Manager, had almost two dozen productive and inspiring business meetings during Taiwanese-Slovak Investment Opportunities Forum in Taipei and individually with companies like Netio Technologies, DataVan International Corporation, A Heng, Argosy Research, Acqutek, and others.

He visited Foxconn (Hon Hai Technology Group), the world’s largest electronics manufacturer.

Another interesting meeting was at Chenbro, which is a pioneer in designing and manufacturing own brand rackmount systems, tower servers, and PC chassis.

Hsinchu Science Park is a smart ecological science park employing 150,000 people. The main sector of the park is the integrated circuit industry including semiconductor foundry TSMC (Taiwan Semiconductor Manufacturing Company).

Robert Cisar met Dr. Wayne Wang, Director General of the Hsinchu Science Park Bureau, and participated in the discussion about potential opportunities for mutual cooperation.

Robert visited also Delta Electronics, one of the most important investors in Slovakia and met R.T. Tsai, who served as Vice President at Delta for more than 40 years.

X Talent Program is a Taiwan high-tech talent circulation program seeking partnerships that advance beyond borders. Robert met Ru-Shin Chen, director of the program, and we are looking forward to welcoming the first talent from Taiwan at Tachyum soon.

Tachyum has completed Floating-Point Unit (FPU) verification for Prodigy, the world’s first Universal Processor, having passed all tests to ensure Prodigy adheres to industry computation standards of Institute of Electrical and Electronic Engineers (IEEE). Prodigy passed the package of tests — more than 24,000,000 in all — in RTL simulation and on Prodigy FPGA prototype hardware.

Tachyum announced its latest milestone of officially entering the final phase of test and development for the Prodigy Universal Processor. This last stage of Quality Assurance (QA) testing of all necessary ported software on a Field Programmable Gate Array (FPGA) for final testing will ensure the chip is production worthy.

Tachyum has entered into an agreement with Senko Advance Co., Ltd. for Optical Interconnect Solution for Prodigy systems enabling high-speed data transfer in data centers. The criterion for selection was the robustness, reliability, and versatility of the SN (Simplified Network) connectivity solutions.

SENKO’s SN connectivity solutions are known for their high data rates and robustness. The SN breakout solution is designed to provide high-speed data transmission at rates of up to 100 Gbps per port. The SN breakout solution is an ideal solution for high performance computing, data center, and cloud applications that require high data rates, low latency, and high reliability.

Tachyum released the second edition of the “Tachyum Prodigy on the Leading Edge of AI Industry Trends” white paper featuring updates such as the implementation of improved 8-bit floating point (FP8) quantization aware techniques with adaptive scaling achieving 32-bit floating point (FP32) accuracy.

In the white paper, Tachyum demonstrates the optimality of the FP8 format for quantizing deep neural networks including weights, activations and gradients, exploiting the fact that floating point numbers provide better coverage than 8-bit integer data type. The results show that FP8 quantized networks can maintain accuracy on par – or even exceed the accuracy – of baseline FP32 models.

“Our experimental results show that FP8 enables faster training and reduced power consumption without any degradation in accuracy for a range of deep learning models. This is one of the most significant AI milestones Tachyum wanted to achieve before tape-out to ensure we have what it takes to make FP8 with sparsity and super-sparsity mainstream AI technology”

Dr. Radoslav Danilák