NOTA Selected as Lead Organization for K-On-Device Humanoid National Strategic Project
Published2026.09.08 08:44
The structure has LG Electronics overseeing the entire project and developing the humanoid system, Mobilint supplying domestic NPUs, and NOTA responsible for AI model optimization software.
The core objective is to lighten the vision language action model (VLA) deployed in humanoids and convert it to match Mobilint NPU's computational architecture, enabling real-time inference even in environments with limited power and computational resources.
Given the characteristics of physical AI where camera imagery must be processed in real-time to enable immediate recognition, judgment, and action, inference latency directly impacts action accuracy and stability, the company explained.
The lightweight model will be converted to match Mobilint NPU architecture, with real-time inference environment and workload-linked modules also being established.
The company will subsequently verify actual performance and resource efficiency in an integrated NPU and humanoid environment.
The company explained that reduced external transmission of sensitive data strengthens security, and improvements in computational efficiency are expected to reduce power consumption.
Related news
.png)
Nota Joins AMD Robotics Partner Network… Expanding On-Device AI Robotics
Nota, a specialist in AI model optimization, has joined the AMD Robotics Partner Network to expand its on-device AI optimization technology into the robotics sector. The company plans to apply model optimization and hardware-tailored optimization techniques to reliably operate AI models in robotic environments where power, memory, and computational resources are limited.
2026.07.24by 배종인 기자
Nota Reduces AI Infrastructure Costs by Up to 76.5%... Unveils Lightweight Model Dedicated to 'Solar Open2'
Nota, a specialized AI model optimization company, has unveiled a lightweight version of Upstage's government-backed AI foundation model Solar Open2. By applying MoE quantization and pruning techniques, the company reduced weight memory by 76.5% and decreased the required GPUs from eight to two. The optimization is expected to deliver significant cost savings for enterprises' AI infrastructure expenses.
2026.07.27by 배종인 기자
Nota, MoonShot AI 'Kimi K3' Lightweight Version…Nvidia B300 GPU Reduced from 8 to 4 Units
AI model optimization company Nota has lightweight-optimized MoonShot AI's 2.8 trillion parameter ultra-large AI model 'Kimi K3'. By applying proprietary non-uniform expert pruning technology, the company maintained core performance while reducing required GPUs from 8 units to 4 units, enabling infrastructure cost savings. The technology analyzes layer-specific characteristics and selectively removes only low-contribution modules, allowing for expected reductions in infrastructure costs.
2026.08.06by 배종인 기자
Nota and MetaM Establish AI Consultation System Development Contract…Entering AI Agent Operating Infrastructure Market
AI model optimization company Nota is making a full-scale entry into the AI agent operating infrastructure market. As its first commercial project, the company is embarking on next-generation AI consultation system development with MetaM, a specialist in AI contact centers (AICC).
2026.08.13by 배종인 기자
Nota, Upstage Consortium Advances to Third Round of Independent Foundation Model
Nota, an AI model optimization and lightweight company, has advanced to the third round of the government's independent AI Foundation Model (DokPaMo) initiative as a member of the Upstage Consortium. Through optimization of Upstage's 'Solar Open 2', Nota enabled a model requiring 8 H100 GPUs to run on 2 units, reducing memory usage by 76.5%, and achieved 80,000 cumulative downloads on Hugging Face. The goal is to enhance industrial applicability of the nation's representative AI models through integration of domestic AI software and hardware ecosystems.
2026.08.20by 배종인 기자
