AI BriefingFuriosaAI, RNGD Sampling Underway
Key point
FuriosaAI has officially begun customer sampling of RNGD, its next-generation chip for data center inference.
Details
FuriosaAI has begun the first PoC (Proof of Concept) for RNGD, its second-generation chip for data center inference. The company recently delivered a server equipped with RNGD to a global tech company based in Seoul, and multiple customers are preparing for sampling to test models.
RNGD is optimized for high-performance, high-efficiency inference for LLMs and multimodal models. It can currently run state-of-the-art models such as Meta's Llama 3.1 70B at full speed through Furiosa's software stack.
RNGD has currently been deployed in the following environments:
- Installed at data centers of the Artificial Intelligence Industry Convergence Agency (AICA) and the National IT Industry Promotion Agency (NIPA) through the K-Cloud project
- A Memorandum of Understanding (MoU) signed with Saudi Arabia's Aramco
Customers will go on to verify real-world performance including latency, throughput, and energy consumption efficiency. FuriosaAI plans to release detailed technical information, including third-party benchmarks, soon.
This summary was generated automatically by AI. Check the original for the author's claims and context. Copyright belongs to the original author.
Our guide explains how the AI works. Report summary errors, attribution issues, or removal requests via Contact.
Related
Qwen3.5-based Small Decision Model 'Kev' Released with TypeSafe API Compatibility and Performance Metrics