FuriosaAI and GUC Partner on RNGD, the Most Efficient AI Accelerator for LLMs
Santa Clara, CA – July 31, 2024 – FuriosaAI Inc. is pleased to announce its collaboration with GUC (Global Unichip Corporation) as a key partner on the second-gen AI accelerator, RNGD (pronounced “Renegade”), the most efficient data center accelerator for high-performance large language and multimodal models.
GUC has been crucial in helping FuriosaAI to achieve an aggressive timeline for RNGD. Developing a new AI accelerator – especially one like RNGD, which utilizes advanced technology like HBM3 and a large-scale 653mm2 SoC with 40 billion transistors – is only possible with excellent partners like GUC, which brings deep expertise in designing system-on-chip (SoC) solutions.
“RNGD offers the unprecedented combination of performance, power efficiency, and programmability for running large language models and multimodal models. GUC's expertise in areas such as SoC design, backend implementation, and silicon interposer (CoWoS®) packaging has ensured that the RNGD project stayed on schedule to launch this year,” said June Paik, FuriosaAI CEO.
Complementing FuriosaAI's cutting-edge technology is GUC's expertise in turnkey semiconductor design services. As a trusted partner, GUC meticulously executed the project starting from spec-in, seamlessly integrating complex HBM memory subsystems, implementing the physical design, designing the package utilizing TSMC’s industry leading CoWoS® technology, followed by comprehensive testing services and manufacturing logistics management, successfully enabling FuriosaAI RNGD’s 1st time success. GUC played a pivotal role in enabling FuriosaAI RNGD's flawless execution, significantly accelerating its time-to-market.
“We are excited to partner with FuriosaAI to bring their revolutionary AI accelerator to market," said Dr. Sean Tai, president of GUC. "GUC is highly experienced with large-scale chip design. Our turnkey design services provide end-to-end support, enabling FuriosaAI to rapidly scale their innovation and meet the growing demand for AI acceleration solutions. We look forward to further supporting FuriosaAI as they continue to design innovative products for the AI market."
Together, FuriosaAI and GUC are poised to drive AI innovation to new heights, empowering organizations worldwide to harness the full potential of artificial intelligence.
For more information about FuriosaAI and the RNGD AI accelerator, visit https://furiosa.ai
For inquiries about GUC’s turnkey design services, visit https://www.guc-asic.com
|
Related News
- RaiderChip unveils its fully Hardware-Based Generative AI Accelerator: The GenAI NPU
- BrainChip Unveils Edge AI Box Partner Ecosystem for Gestures, Cybersecurity, Image Recognition, and Computer Vision
- BOS and Tenstorrent Unveil Eagle-N, Industry's First Automotive AI Accelerator Chiplet SoC
- Synopsys Announces Industry's First Ultra Ethernet and UALink IP Solutions to Connect Massive AI Accelerator Clusters
- Arteris and MIPS Partner on High-Performance RISC-V SoCs for Automotive, Datacenter and Edge AI
Breaking News
- Breker RISC-V SystemVIP Deployed across 15 Commercial RISC-V Projects for Advanced Core and SoC Verification
- Veriest Solutions Strengthens North American Presence at DVCon US 2025
- Intel in advanced talks to sell Altera to Silverlake
- Logic Fruit Technologies to Showcase Innovations at Embedded World Europe 2025
- S2C Teams Up with Arm, Xylon, and ZC Technology to Drive Software-Defined Vehicle Evolution
Most Popular
- Intel in advanced talks to sell Altera to Silverlake
- Arteris Revolutionizes Semiconductor Design with FlexGen - Smart Network-on-Chip IP Delivering Unprecedented Productivity Improvements and Quality of Results
- RaiderChip NPU for LLM at the Edge supports DeepSeek-R1 reasoning models
- YorChip announces Low latency 100G ULTRA Ethernet ready MAC/PCS IP for Edge AI
- AccelerComm® announces 5G NR NTN Physical Layer Solution that delivers over 6Gbps, 128 beams and 4,096 user connections per chipset
![]() |
E-mail This Article | ![]() |
![]() |
Printer-Friendly Page |