Annonce • Jun 13
Nota Ai Has Two Moe Quantization Papers Accepted At Icml 2026 Workshop Nota AI announced that two of its papers on MoE-specific quantization algorithms have been accepted to the Resource-Adaptive Foundation Model Inference (AdaptFM) Workshop at ICML 2026. The AdaptFM Workshop focuses on technologies that enable large-scale AI models to run efficiently under limited computing resources. This achievement is significant as it recognizes Nota AI's accumulated technical expertise in optimizing Mixture-of-Experts (MoE) models, an architecture increasingly regarded as a core structure for large language models (LLMs). MoE models improve both performance and efficiency by activating only a subset of expert models as needed. However, their complex structure requires a different approach to quantization, the process of making models smaller and more efficient, compared to conventional model architectures. Nota AI previously won both its track and the overall competition at the NVIDIA Nemotron Hackathon with a data-driven MoE quantization method. With the acceptance of these two papers, Nota AI will once again present research outcomes specifically designed for MoE architectures on a global research stage. The first accepted paper, "DREAM-MoE," proposes a method to reduce changes in a model's decision flow that can occur when large-scale AI models are quantized across multiple segments. The method focuses on the fact that even a small error in an earlier segment can affect expert selection in later segments. DREAM-MoE helps the quantized model select experts in a way that remains closer to the original model. The second paper, "SRA-MoE," proposes a method that identifies and prioritizes important inputs that have a greater impact on the model's final output. Rather than treating all inputs equally, SRA-MoE is designed to prevent expert selection from being significantly disrupted for these key inputs, helping maintain model quality more effectively under limited resources. Both studies demonstrated higher performance compared to the latest MoE-specific quantization methods. This shows that large-scale AI models can be executed with less memory and fewer computing resources while reducing quality degradation. Nota AI has been proactively focusing its R&D efforts on optimizing large AI models that require substantial memory and computing resources. The company is advancing large-scale model optimization, including Solar MoE, as part of the sovereign foundation model project led by the Upstage consortium. It is also expanding its experience in quantizing NVIDIA Nemotron 3 Nano to newer large models such as Nemotron Ultra, further broadening the scope of its optimization technologies. In addition, Nota AI will host "Nota AI - Korea Efficient Days" during ICML 2026 at COEX in Seoul. The event will bring together global researchers, engineers, and business leaders visiting Korea to share research trends and industrial applications of Efficient AI. Through the event, Nota AI plans to introduce its research achievements in large-scale AI model optimization and expand opportunities for technical collaboration and business engagement. New Risk • Jun 06
New major risk - Share price stability The company's share price has been highly volatile over the past 3 months. It is more volatile than 90% of South Korean stocks, typically moving 16% a week. This is considered a major risk. Share price volatility increases the risk of potential losses in the short-term as the stock tends to have larger drops in price more frequently than other stocks. It may also indicate the stock is highly sensitive to market conditions or economic conditions rather than being sensitive to its own business performance, which may also be inconsistent. This is currently the only risk that has been identified for the company. Reported Earnings • May 21
First quarter 2026 earnings released: ₩188 loss per share (vs ₩573 loss in 1Q 2025) First quarter 2026 results: ₩188 loss per share (improved from ₩573 loss in 1Q 2025). Revenue: ₩3.58b (up ₩3.52b from 1Q 2025). Net loss: ₩4.02b (loss narrowed 29% from 1Q 2025). Revenue is forecast to grow 30% p.a. on average during the next 3 years, compared to a 16% growth forecast for the Software industry in South Korea. Annonce • Mar 17
Nota Inc., Annual General Meeting, Mar 31, 2026 Nota Inc., Annual General Meeting, Mar 31, 2026, at 10:00 Tokyo Standard Time. Location: conference room, 1, expo-ro, yuseong-gu, daejeon South Korea Annonce • Mar 10
Nota AI Showcases End-To-End On-Device AI At Embedded World 2026 Nota AI, an AI optimization technology company, announced that it will participate in Embedded World 2026, taking place March 10-12 in Nuremberg, Germany. At the event, the company will present the full lifecycle of on-device AI-from model optimization to deployment in real-world industrial environments. Nota AI will showcase how AI models are optimized through NetsPresso® and deployed across a wide range of global hardware platforms before being implemented in real industrial environments. Nota AI will demonstrate how semiconductor companies can rapidly optimize high-performance AI models for their chips using its AI model optimization platform, NetsPresso®. The company has accumulated extensive expertise in lightweighting and optimizing AI models—from small language models (SLMs) to large language models (LLMs) and vision-language models (VLMs). To date, Nota AI has successfully compressed more than 40 AI models while maintaining performance and has deployed its optimization technologies across over 100 hardware devices. The company recently supplied AI optimization technology for Samsung Electronics' Exynos 2600, where the technology serves as a core component enabling mobile on-device AI capabilities. Nota AI has also maintained ongoing technology collaborations with global semiconductor companies including Qualcomm and Arm. At Embedded World, the company will present live demonstrations showing both computer vision models and large language models running in real time on these hardware platforms, highlighting AI performance in edge environments. Nota AI will also showcase a Device Farm, featuring a collection of hardware platforms optimized by the company over the past decade. Visitors will be able to explore a range of chipsets from major global semiconductor companies running AI models optimized with Nota AI's technology, demonstrating the company's experience in optimizing more than 100 hardware platforms over the past ten years. Nota AI will introduce real-world solutions that combine AI models with hardware optimization in on-device environments. Through its video analytics solution NVA (Nota Vision Agent), Nota AI has delivered technologies across industries such as safety monitoring, security, and smart city operations in collaboration with global partners including NVIDIA. At the booth, the company will demonstrate real deployment cases including selective video monitoring, intelligent transportation systems (ITS), and industrial safety monitoring. Nota AI will also present its latest research achievements recently accepted at ICLR 2026 and the AAAI 2026 Foundation Model Workshop. Both studies focus on improving the efficiency and reliability of vision-language models (VLMs), highlighting Nota AI's technological capabilities across the broader physical AI landscape-from vision-language models to vision-language-action (VLA) systems. During the exhibition, Tae-Ho Kim, CTO & Co-Founder of Nota AI, will host mini sessions at the booth to present the company's AI lightweighting and optimization strategies and share real-world cases of applying Nota AI technologies to global semiconductor platforms. Nota AI will offer complimentary Embedded World visitor passes to attendees who pre-register through the company's official website and visit the Nota booth (Hall 5, Booth 5-422). Annonce • Mar 06
Nota AI Announces Proprietary Quantization Technology For Upstage Solar LLM Nota AI, an AI optimization technology company behind the Nota AI brand, announced that it has developed a next-generation quantization technology that significantly compresses the size of Solar, a high-performance large language model (LLM) developed by Upstage, while maintaining high accuracy. The breakthrough reduces inference costs and improves processing speed without sacrificing performance. The development was carried out as part of the Sovereign AI Foundation Model Project led by South Korea's Ministry of Science and ICT. By applying Nota AI's lightweighting and optimization technologies to Solar Open 100B, the company significantly improved memory efficiency while preserving model performance. The achievement lowers the memory requirements of the 100B-parameter model while maintaining its capabilities, enabling more practical deployment of Korean AI foundation models in physical AI environments such as mobility and robotics. The newly developed technology focuses on addressing technical challenges associated with the Mixture of Experts (MoE) architecture, which is rapidly gaining adoption in next-generation LLMs. Conventional quantization methods typically compress the entire model uniformly without considering the distinct characteristics of individual expert models. To overcome this limitation, Nota AI developed a proprietary algorithm optimized for MoE architectures, called Nota AI MoE Quantization. The approach is designed to minimize quantization distortion during the inference process of MoE models. Unlike conventional methods that uniformly reduce precision across all operations, Nota AI's algorithm selectively preserves precision in critical components while compressing less sensitive parts of the model. This enables effective model compression while minimizing performance loss. Applying the technology to the Solar 100B model yielded significant improvements compared with conventional quantization methods. Nota AI successfully reduced Solar's memory usage from 191.2GB to 51.9GB, representing a 72.8% reduction. At the same time, the model maintained performance levels comparable to the original version, achieving a Perplexity (PPL) score of 6.81, close to the baseline model's 6.06. In contrast, some generic quantization approaches resulted in performance degradation exceeding fivefold. Nota AI has filed a patent application for the technology to strengthen its intellectual property portfolio. While conventional quantization techniques often sacrifice model performance to reduce memory usage, Nota AI's technology demonstrates that it is possible to maintain performance while delivering AI services faster and to more users on limited GPU infrastructure. As a result, enterprises can deploy large-scale LLMs more easily on their own devices—models that were previously difficult to implement due to hardware constraints. The significant reduction in Solar 100B's memory footprint while preserving performance also creates new opportunities for deploying high-performance AI in real-world on-device environments, including robotics and automotive systems. Additionally, the technology enables organizations facing limited access to high-end GPU infrastructure to serve more users on the same hardware, directly contributing to lower operational costs. 