Connect with us

Technology

Skymizer Launches Groundbreaking LLM Accelerator IP for on-device LLM Inferencing, EdgeThought, the game-changer in on-device GenAI era.

Published

on

TAIPEI, May 31, 2024 /PRNewswire/ — Skymizer, a pioneer in compiler technology and optimized solutions, today announced the release of its revolutionary software-hardware co-design AI ASIC IP, EdgeThought, specifically engineered for accelerating Large Language Models (LLMs) at the edge. This cutting-edge innovation leverages Skymizer’s advanced compiler technology to set new industry benchmarks in computation, memory utilization, power efficiency, and cost-effectiveness.

EdgeThought is designed to enhance the performance of LLM applications in a wide range of edge devices, from IoT devices to automotive systems and AI PCs / AI Edge Servers etc. By optimizing EdgeThought design with the compiler-centric approach, Skymizer’s solution ensures that these devices can run the state-of-the-art on-device LLM models, including the newest Llama3 8B.

Key Features of Skymizer’s LLM Accelerator:

Optimized Compiler Technology: At the heart of the accelerator is Skymizer’s proprietary compiler technology, which maximizes hardware utilization and efficiency, enabling superior performance even on resource-constrained edge devices.Enhanced Computation and Memory Efficiency: The co-design approach minimizes latency and maximizes throughput while reducing memory footprint, allowing for faster and more reliable inferencing of LLMs at the edge.Power and Cost Efficiency: Skymizer’s solution drastically reduces the power consumption and operational costs associated with deploying advanced AI models, making it an ideal choice for businesses looking to scale their operations sustainably.Scalability and Flexibility: Designed to support a range of LLM applications, the accelerator is scalable to different sizes and performance requirements, including multiple users with multiple batches to scale up the throughput for a very power edge server, offering unprecedented flexibility for device manufacturers and application developers.

“Today marks a significant milestone not just for Skymizer, but for the entire AI and edge computing industries,” said Jim Lai, CEO of Skymizer. “Our innovative LLM accelerator redefines what’s possible in edge AI performance, making it both more accessible and cost-effective. This release reflects our commitment to pushing the boundaries of technology to empower our clients and enrich the user experiences.”

With a decade of compiler and virtualization industry experience Skymizer focuses on what it does the best and designs EdgeThought to be a compiler optimized LLM ASIC IP specific for on-device inferencing task. EdgeThought eliminated all the software and hardware needs for training and focus on inferencing only. The result is the best in class on-device LLM inference engine. “If Groq chip is the king of cloud LLM inferencing, then, EdgeThought will be the game-changer for on-device LLM inferencing.”, said William Wei, the CMO & EVP of Skymizer. “And, EdgeThought does not require the latest silicon manufacturing process, it can enable the less expensive mature silicon class and specialty memory components which will revitalize those cheaper memory industry in the GenAI era.”

EdgeThought IP is now available for licensing and can be integrated into existing hardware setups without the need for extensive modifications. Skymizer has already begun collaborating with leading semiconductor companies and device manufacturers to bring this powerful technology to market.

About Skymizer:

Skymizer is a global leader in compiler technology and AI optimization solutions. Based in Taipei, Taiwan, the company is dedicated to innovating how software and hardware interact to create more efficient, powerful, and cost-effective computing solutions. Skymizer’s products are used in a variety of applications across multiple industries, driving the evolution of computing at the edge.

For more information on Product:
Website: www.skymizer.ai

Note: This press release is for informational purposes only and does not constitute an offer to sell, a solicitation to buy, or a recommendation for any security or product, nor does it constitute an offer to provide investment advisory or other services by Skymizer.

View original content:https://www.prnewswire.com/news-releases/skymizer-launches-groundbreaking-llm-accelerator-ip-for-on-device-llm-inferencing-edgethought-the-game-changer-in-on-device-genai-era-302160453.html

SOURCE Skymizer

Continue Reading
Click to comment

Leave a Reply

Your email address will not be published. Required fields are marked *

Technology

Bloomberg Introduces Spread-to-Benchmark Quoting for EUR and GBP Portfolio Trading Baskets

Published

on

By

LONDON, June 2, 2026 /PRNewswire/ — Bloomberg today announced the launch of Spread-to-Benchmark quoting and trading for Euro (EUR) and Sterling (GBP) denominated portfolio trades through its Portfolio Trading Basket Builder (PTBB). The new functionality expands the range of quoting protocols available for European credit portfolio trading and reflects growing client demand for spread-based execution workflows, alongside increased dealer support for the convention across EUR and GBP markets.

Spread-to-Benchmark quoting is a well-established protocol for USD credit portfolio trades and is used by market participants to evaluate and execute portfolio trades. By extending this workflow to EUR and GBP portfolio trades, Bloomberg enables clients and dealers to transact using a familiar spread-based methodology across additional credit markets. 

The introduction of Spread-to-Benchmark quoting for EUR and GBP baskets reflects increased client interest in evaluating portfolio trades through a spread-based lens and the growing adoption of spread-based execution workflows in European credit markets. The workflow provides market participants with an additional framework for assessing the relationship between credit spread risk and underlying government bond yields when pricing and executing portfolio trades. 

Additional Workflow Flexibility 
The workflow complements Bloomberg’s existing portfolio trading capabilities, which support the full range of market-standard quoting conventions, including Price, Yield, Spread-to-Benchmark and Spread based workflows that reference Bloomberg’s evaluated pricing service (BVAL). This gives clients flexibility to compare and execute portfolio trades using the quoting methodology that best aligns with their investment objectives, execution preferences and internal risk management processes. 

“European credit clients continue to look for execution workflows that reflect how they evaluate risk and monitor portfolio trading outcomes,” said Harry Street, Global Head of Credit and Equities Trading Product at Bloomberg. “By expanding dealer support for Spread-to-Benchmark quoting for EUR and GBP baskets, Bloomberg is broadening the range of workflow options available to clients trading European credit portfolios.” 

“Portfolio trading workflows in fixed income continue to become more sophisticated as institutional investors look for ways to evaluate execution quality in changing market conditions,” said Kevin McPartland, Head of Market Structure & Technology Research at Crisil Coalition Greenwich. “Spread-based quoting helps market participants more clearly distinguish between the impacts of credit spread and underlying rates movements when determining how best to execute a portfolio trade.” 

Bloomberg’s Electronic Markets solutions are used by leading financial institutions to trade efficiently in over 175 markets around the world. More than 9,000 client firms use Bloomberg Electronic Markets to access industry-leading depth and breadth of liquidity across asset classes from over 800 dealers globally. Bloomberg Electronic Markets provides market participants with comprehensive solutions across the trading lifecycle, including robust price transparency, analytics, automation and execution, powered by Bloomberg’s high-quality, multi-asset class data and tools.

About Bloomberg
Bloomberg is a global leader in business and financial information, delivering trusted data, news, and insights that bring transparency, efficiency, and fairness to markets. The company helps connect influential communities across the global financial ecosystem via reliable technology solutions that enable our customers to make more informed decisions and foster better collaboration. For more information, visit Bloomberg.com/company or request a demo.

View original content to download multimedia:https://www.prnewswire.com/news-releases/bloomberg-introduces-spread-to-benchmark-quoting-for-eur-and-gbp-portfolio-trading-baskets-302787922.html

SOURCE Bloomberg L.P.

Continue Reading

Technology

Dr. Sunho Kang, a senior battery-technology executive with leadership experience at major global battery and EV manufacturers, joins TeraWatt Technology as Head of Product and Technology

Published

on

By

SAN FRANCISCO, June 2, 2026 /PRNewswire/ — TeraWatt Technology Inc. (Headquartered in California, USA) is pleased to announce that Dr. Sunho Kang has joined the company as Head of Product and Technology.

Dr. Kang is a globally recognized battery-technology executive with more than 25 years of leadership experience spanning the United States, Asia, and Europe, and a distinguished track record of advancing innovations from laboratory research through gigafactory-scale production. He has held senior executive positions at world-leading organizations including Samsung SDI, Apple, and Volkswagen Group of America, and brings deep expertise in lithium-ion battery materials, cell engineering, and product industrialization across a broad range of applications, including electric vehicles and energy storage systems.

At TeraWatt, Dr. Kang will lead global product development and the commercialization of TeraWatt’s battery technology platform, aiming to accelerate the delivery of TeraWatt’s competitive products as well as the technology and commercialization roadmap including manufacturing scale-up.

Dr. Kang commented:

“I am thrilled to join TeraWatt Technology as Head of Product and Technology. TeraWatt’s innovative battery platform presents a tremendous opportunity to push the boundaries of lithium-ion technology, and I look forward to working with the team to accelerate product development and commercialization to deliver meaningful impact.”

TeraWatt Technology founder CEO Ken Ogata, Ph.D. commented:

“We are thrilled to welcome Dr. Kang as our Head of Product and Technology. His deep expertise in battery materials, cell engineering, and productization will be instrumental in accelerating TeraWatt’s product roadmap and technology leadership. Together with Dr. Kang, we will continue to drive our mission forward.”

About TeraWatt Technology Inc.
TeraWatt Technology Inc. is a California-based company that produces lightweight, high-power, and safe next-generation lithium-ion batteries.

Company Overview
Name: TeraWatt Technology Inc.
Representative: Co-founder and CEO Ken Ogata
Headquarters: 28 Geary St, Suite 650, San Francisco, CA 94108, United States
Founded: January 2020
Established: December 2019
URL: https://www.terawatt-technology.com/

 

View original content:https://www.prnewswire.com/news-releases/dr-sunho-kang-a-senior-battery-technology-executive-with-leadership-experience-at-major-global-battery-and-ev-manufacturers-joins-terawatt-technology-as-head-of-product-and-technology-302786854.html

SOURCE TeraWatt Technology Inc.

Continue Reading

Technology

Tencent Cloud and Soniox Announce Strategic Partnership: Combining Advanced Speech-to-Text (STT) Technology with Global Real-Time Infrastructure

Published

on

By

HONG KONG, June 2, 2026 /PRNewswire/ — Tencent Cloud, the cloud business of global technology company Tencent, today announced a strategic partnership with Soniox, a San Francisco-based speech AI company that specializes in developing high-accuracy, low-latency speech AI solutions. The collaboration integrates Soniox’s speech-to-text (STT) technology with Tencent Cloud’s Real-Time Communication (TRTC) enterprise-grade global infrastructure, enabling enterprises to build and deploy multilingual voice AI applications across 200+ countries and regions.

Elevating Enterprise Voice AI at a Global Scale

In enterprise voice AI deployments, latency directly affects user experience and application reliability. The integration of Soniox’s high-accuracy, low-latency STT with TRTC’s global transmission infrastructure reduces latency across the entire pipeline, creating a comprehensive end-to-end solution for enterprises deploying conversational AI applications worldwide.    

Soniox is the voice platform for every language. Unlike legacy speech AI, which was built primarily for English-speakers, Soniox delivers native-speaker accuracy across 60+ languages. Its technology can handle mid-sentence language switching — a user can switch between English and Chinese in a single utterance, and Soniox will capture every word with complete accuracy. All of this works through a single API that works for both speech-to-text and text-to-speech.

By integrating TRTC, the partnership leverages an enterprise-grade real-time communication backbone featuring more than 3,200 global nodes, sub-300 ms worldwide latency, and advanced capabilities such as AI noise suppression and weak-network resilience. These capabilities enable conversational AI applications to operate reliably across diverse network environments, including regions such as Southeast Asia and Africa.

With the roll out of this partnership, developers can integrate the Soniox STT API directly within the Tencent Cloud console. Whether targeting English-speaking markets or supporting languages such as Arabic, Hindi, and Malay, enterprises can build global voice applications — including intelligent customer service, voice assistants, real-time translation, and meeting transcription — to address the demands of expansion into emerging markets and multilingual scenarios.

Wison Xie, Head of Product at Tencent RTC, stated: “Tencent RTC has always been committed to providing reliable real-time communication infrastructure for global enterprises. Our partnership with Soniox brings together our strengths in enterprise-grade audio transmission and Soniox’s advanced speech recognition technology. Together, we are making it easier for businesses to deploy accurate, low-latency voice AI applications across any language and any market.”

Klemen Simonic, CEO at Soniox Inc., stated “At Soniox, our mission is to help businesses understand every word, in any language, with native speaker accuracy and exceptional speed. Partnering with Tencent Cloud combines our speech AI with world-class real-time infrastructure, enabling enterprises to build voice AI experiences that scale globally with low latency and reliability.”

About Tencent Cloud:

Tencent Cloud, one of the world’s leading cloud companies, is committed to creating innovative solutions to resolve real-world issues and enabling digital transformation for smart industries. Through our extensive global infrastructure, Tencent Cloud provides businesses across the globe with stable and secure industry-leading cloud products and services, leveraging technological advancements such as cloud computing, Big Data analytics, AI, IoT, and network security. It is our constant mission to meet the needs of industries across the board, including the fields of gaming, media and entertainment, finance, healthcare, property, retail, travel, and transportation.

About Tencent RTC:

Tencent RTC provides real-time communication solutions, including audio/video calling, live streaming, and in-game voice. With enterprise-grade security, AI-powered enhancements, and a global network of over 3,200 nodes, Tencent RTC powers mission-critical communication for customers worldwide.

About Soniox:

Soniox is a next-generation voice AI company bringing about the end of English-first speech AI. Most people on the planet did not grow up speaking English and often mix languages mid-sentence; and yet legacy speech AI was built for just English. Soniox is different: native-speaker accuracy across 60+ languages, true mid-sentence language switching, and flawless alphanumeric recognition that legacy providers still can’t match. For developers building global apps, Soniox is the only option. Try it for yourself at soniox.com.

View original content to download multimedia:https://www.prnewswire.com/apac/news-releases/tencent-cloud-and-soniox-announce-strategic-partnership-combining-advanced-speech-to-text-stt-technology-with-global-real-time-infrastructure-302786832.html

SOURCE Tencent Cloud

Continue Reading

Trending