NVIDIA Feynman Aiming for Mass Production in the Second Half of 2028
NVIDIA's next-generation AI platform, Feynman, is being developed with the goal of mass production in the second half of 2028. TSMC's A16 process, 3D stacking, and chip-on-wafer packaging (CPO) technologies will be introduced. TSMC has adjusted its monthly production capacity for SoIC from 20,000 wafers at the end of 2026 to 50,000 wafers at the end of 2027. The A16 aims to increase speed by 8-10% or reduce power consumption by 15-20% at the same voltage by combining nanosheet transistors with a rear power delivery network. Feynman will require a more complex packaging structure than the Rubin generation, incorporating 3D chiplets, SoIC, and CPO technologies. NVIDIA announced at GTC 2026 that the Feynman generation will include new GPUs, CPUs, BlueField 5, and CX10. The NVLink interconnect bandwidth of the Feynman platform is expected to exceed 1PB/s, which is higher than the 520TB/s of Rubin Ultra. NVIDIA plans to invest over $40 billion in the AI ecosystem after 2026. TSMC's expansions of AP7 and AP8 are also part of a roadmap targeting the demand for next-generation AI semiconductors.
-- Price
This content is provided for general informational purposes only and doesn't constitute financial, investment, legal, or tax advice. Any events, rewards, online promotions, or related information mentioned herein should not be considered a recommendation, solicitation, or invitation to purchase, sell, trade, or otherwise deal in any crypto assets. Crypto assets are highly volatile and may result in loss. The availability of WEEX services, products, and related events may vary by region. You are responsible for ensuring that your participation is in accordance with applicable local laws and regulations.
You may also like

Apple Plans Global Price Adjustment for iPhone 17 Series, Individual Price Increase May Reach 1000 Yuan

NVIDIA Expands AI Financing Plan to $500 Billion

Mark Cuban Predicts Chips Will Become the New Crypto Asset, Will This Time Fail?

Jensen Huang, Ultraman, and Masayoshi Son: A 20-Year Alliance

Arm Holdings Acknowledges Review of Acquisition of Cerebras and Marvell

web3: Coldcard Seed Vulnerability Exposed, Over $100 Million in Bitcoin Stolen

AI Storage Chips Strengthen, SK Hynix Continues Pre-Market Rise

Groq Raises $350 Million, Valuation Drops to $3.5 Billion

Financialization of AI Computing Power: Open Source Models are Pushing Computing Power to Capital Markets (Part 2)

CZ's One-Click Triple Action Boosts 16-Year-Old's On-Chain CPU Project

NVIDIA Plans to Invest $3 Billion in SB Energy to Support OpenAI Project

U.S. Stocks Hit Record Highs, Korean Stocks Return to Technical Bull Market, Market Focus on Three Key Variables

Mark Cuban Sees AI Chips as a New Investment Frontier

Taiwan's GDP Grows 13.72% in First Half: AI Export Boom Driving Growth to 40-Year High

Multiple US ETFs Increase Holdings in Chinese Chip Stocks

Short Selling in South Korea Rises as Morgan Stanley Increases BTC Holdings by Over 6,600 Coins

$500 Billion AI Finance: Off-Balance Sheet Debt Grows

Rubin Ultra Average Selling Price Nears $170,000, Memory Costs Account for Half

Lumentum Reports 109% Increase in Q4 Revenue Amid Ongoing Optical Chip Supply Issues

DeepX Secures 77 Orders for On-Device AI Chip DX-M1 One Year After Mass Production

The Return of NFT Summer: Which Projects on Robinhood Chain Deserve Attention?

web3: OpenAI Launches Ultrafast Mode, GPT 5.6 Sol Speed Increased to 14 Times

AMD Plans to Issue $5 Billion in Bonds to Boost AI Infrastructure

The Revolution of Humanoid Robots: When Will They Arrive in Homes and Where Do Investors See Opportunities

Bank of America Commits $250 Billion to U.S. Infrastructure

Lenovo Plans to Launch NVIDIA RTX AI PCs in the Second Half of the Year

BTC Short-Term Holder Cost Line Under Pressure for Nearly 2 Months

Capital Economics: South Korea's Semiconductor Boom May Cool Within Two Years

KOSPI Index Enters Technical Bull Market as AI Boom Fuels Rebound



