Nota AI

Team

company

Verified

https://www.nota.ai/

nota_ai

nota-github

Activity Feed Request to join this org

AI & ML interests

Hardware-aware AI Model Optimization

Recent Activity

SangminLee-NOTA updated a Space about 1 hour ago

nota-ai/README

bokyeong1015 updated a collection 2 days ago

Efficient MoE-based LLM

hancheolp new activity 2 days ago

nota-ai/Solar-Open-100B-Nota-FP8:Please make nvfp4 quant and make mlx version of that quant

View all activity

Organization Card

Community About org cards

Nota AI bridges the gap between high-performance AI models and edge devices.
From our automated optimization platform to bespoke AI solutions, we ensure your AI functions efficiently—everywhere it is needed.

🌟 Spotlight

Sovereign AI Foundation Model Project

Nota AI participates in the Sovereign AI Foundation Model Project, a key initiative by the South Korean government (NIPA) to develop global-tier foundation models. As a core optimization partner, we focus on compressing massive LLMs for practical deployment.

Optimized Solar-Open-100B Series

🏆 Official Quantized Model: Solar-Open-100B-NotaMoEQuant-Int4

This is the official quantized version of Upstage flagship model, Solar-Open-100B. It is specifically optimized to run on a single A100 (80GB) GPU.

Why NotaMoEQuant: Our approach directly resolves representational distortion in MoE architectures, delivering superior benchmark accuracy compared to conventional methods.

Hardware Efficiency: While the original model required four A100 (80GB) GPUs, our optimized version reduces this requirement to just a single A100 (80GB) for standard inference. Only two A100 GPUs are needed for maximum context generation.

🔥 New Release: Solar-Open-100B-NotaMoEQuant-NVFP4

This latest release utilizes NVFP4 quantization and compressed-tensors packing to ensure backend compatibility with Hugging Face and vLLM. It is specifically designed for NVIDIA Blackwell architecture, requiring a minimum of one B100 GPU.

Also available: Solar-Open-100B-Nota-FP8

🚀 Our Core Business

🛠️ AI Platform: NetsPresso

"We make AI lighter, faster, and ready for deployment."

NetsPresso is our proprietary platform that accelerates model optimization, enabling you to secure on-device latency and accuracy without deep hardware expertise.

Develop & Compress: Create lightweight models effortlessly using our Model Zoo and advanced Compressor (Structured Pruning).
Optimize & Convert: Maximize speed on verified hardware (NVIDIA, Arm, Qualcomm, etc.) with Graph Optimization and Graph Quantization.
Test on Real Devices: Validate performance instantly on actual devices via our Device Farm to eliminate deployment failures.

👉 Ready to optimize? Try NetsPresso Now | View Documentation

🌍 AI Solutions

"We provide end-to-end AI solutions powered by our core optimization technology."

1. Nota Vision Agent

Powered by Vision Language Models (VLM), this agent goes beyond simple detection to understand complex situational contexts.
It interprets video feeds through natural language prompts, delivering real-time insights locally without cloud dependency.

2. Edge AI Solutions

🚦 Intelligent Transportation Systems (ITS): We optimize real-time traffic control and analysis to enhance urban mobility.
🚗 Driver Monitoring System (DMS): We provide lightweight AI for driver safety, ensuring fast and accurate facial recognition inside vehicles.
🛡️ Security Surveillance: We deliver efficient video analytics designed to maintain public and industrial safety with minimal resource usage.

📚 Tech Blog

Gain insights into our engineering philosophy. We share deep dives into model compression methodologies and NPU acceleration techniques to help you stay ahead.

👉 Read our Tech Blog