AI BriefingRebellions news
The latest AI and developer news from Rebellions, with short summaries.
Visit website →ReQAT Achieves BF16 Accuracy with 4-bit Quantization for Weights, Activations, and KV Cache
The method identifies that FP4 errors stem from low-entropy tokens and uses trace-aligned training to reach 65.94% AIME-120 accuracy on R1-Qwen-14B.
Rebellions Engineering: Switchless Dragonfly and 6D Hypercube Topologies Offer Cost-Effective 64-Device AI Racks
Rebellions AI offers switchless Dragonfly and 6D hypercube topologies for Rebel 100s clusters, reducing capital and power costs while maintaining competitive performance for large-scale GenAI inference.
Rebellions details Rebel100s interconnect hierarchy for AI inference
The Rebel100s accelerator integrates switch functionality into I/O chiplets to support up to 256 sockets without external switches.
RAISE Summit 2026 Keynote | Where the Most Profitable AI Inference Chip Runs
Rebellions presented its AI inference chip strategy at RAISE Summit 2026.
AI Full-Stack Solution Based on Domestic Technology for Patent Technical Document Classification and Summarization
Wert Intelligence combined Rebellions' NPU with PlutoLM to build a domestic AI full-stack solution for patent data classification and summarization.
Physical AI Based on NPU Servers: UAE Water Purification Robot Solution
Rebellions ATOM NPU servers doubled the inference efficiency of a UAE water purification robot.
2026 vLLM Korea Meetup
At the Seoul meetup, participants confirmed the trend of vLLM growing into standard infrastructure for LLM serving.
4-Chiplet AI SoC With 16Gbps UCIe-Advanced Die-to-Die Interface and Full-Chip Scalable Mesh for Large-Scale AI Inference
A 4-chiplet AI SoC implements a full-chip mesh with 16Gbps UCIe-Advanced.
AI HS Code Recommendation Chatbot for Mongolia Customs: Based on LLM and RAG
Mongolia Customs automated HS Code classification with an LLM/RAG-based NPU chatbot.
AI-Powered Veterinary Imaging Diagnosis Support Driven by NPU Infrastructure
SK Telecom's X Caliber will drop GPUs for ATOM™ NPU in 2025.
vLLM Hands-on Workshop Recap
Rebellions and SqueezeBits held a hands-on workshop where participants ran vLLM directly on NPUs.
Innovations in Data Center Compute and Power Supply for Large-Scale AI Sustainability
NPUs and VIB ESS dramatically cut AI data center power costs and peak loads.
The First vLLM Meetup Held in Korea
This is an event page reporting on the first vLLM meetup held in Korea.
REBEL-Quad Live | World's First UCIe-Advanced NPU Demonstrated at Hot Chips 2025
At Hot Chips 2025, REBEL-Quad's world's first UCIe-Advanced NPU demonstration was unveiled on-site.
Integrating LLM-Based AI into SOC Threat Detection and Response
IGLOO implemented the on-premise LLM security assistant AiR using Rebellions NPUs.
NPU-Based LLM Serving: A Redesign for Maximum Scalability and Efficiency
Rebellions unveiled a high-efficiency LLM serving technology compatible with vLLM by optimizing FlashAttention and other techniques for NPUs.
AI-Powered Safety Management for Construction and Plant Sites
KOLON BENIT has launched an AI vision safety management solution for construction sites that maximizes power efficiency using Rebellions' NPU.