Deep Research2
🔒 Members only- 2026.06.29📄 Investment memo
NVIDIA: $4.7T Market Cap, 22x PE, 85% Growth – The AI Core Holding
Buy-side institutional-grade deep-dive memo: World's largest market cap, forward PE just 22x – a comprehensive breakdown of growth, valuation, scenarios, and risks. Members only.
Read memo→ - 2026.06.29📊 Model · .xlsx download
NVDA Financial Model: Assumptions, Forecasts & DCF
Downloadable financial model for NVIDIA (NVDA): core assumptions, financial forecasts, DCF/scenario-based price targets — a quantitative modeling worksheet for subscribers to download and build their own.
Open model→
Coverage Timeline4
Following since 2024.10- 20262026.06.296 min
Nvidia at 22x: The Cheapest Megacap or a Cycle Peak in Disguise?
Revenue still +85%, gross margin 75%, quarterly FCF $48.6B — but it remains a cyclical stock
- 2026.03.174 min
GTC 2026: Nvidia Is the AI Infrastructure General Contractor and the Future Token Factory
1. The trillion-dollar order pipeline: Huang's biggest bombshell—pushing data center revenue visibility to a staggering $1 trillion—completely reshapes market expectations.
- 20252025.10.296 min
Nvidia's GTC Lays Out a Grand Vision for AI's Future
This GTC keynote is a must-watch. Jensen Huang's message: Nvidia has the ambition and capability to become the absolute center of future AI development. AI infrastructure is not a bubble because demand isn't—it's driven by real enterprise cost savings and profit improvements.
- 20242024.10.065 min
Tech Stocks & AI: The Biggest Risks and How to Play Them
This article draws on two Goldman Sachs research reports on tech stocks and AI investing. For accuracy and compatibility, cited content is paraphrased. The original reports can be accessed at the bottom. An English version will also be published on Seeking Alpha.
Understand the industry behind this company — visual, step-by-step explainers.
AI 产业链全景
算力如何从一粒沙,层层传导成你看到的智能。
数据中心:装算力的工厂
机柜、供电、散热、网络——AI 工厂的四大系统。
芯片制造:把蓝图变成硅片
人类最精密的工厂,和一片晶圆的三个月。
模型层:算力变成智能
训练像上学,推理像上班——两种烧钱姿势。
光模块与网络:算力的神经
把一万张 GPU 连成一台机器的那张网。
GPU:AI 的发动机
几万个小核并排干活,一台吞吐数字的机器。
HBM:堆起来的内存
GPU 旁边那几颗小芯片,决定了算力能跑多快。
电力:算力的新瓶颈
从电网到芯片,电走过的路和它的账单。
服务器整机:把芯片装进机器
价值九成在 GPU 托盘,生意九成在集成与交付。