NVIDIA Groq 3 LPX Expected to Ship in the Second Half of the Year
Prefer us on Google
NVIDIA has announced the Groq 3 LPX, an artificial intelligence inference accelerator, as a low-latency processing device for the Vera Rubin platform. This device features a structure with 256 LPUs and 128GB of on-chip SRAM. In a press release on March 16, it was stated that the seven chips constituting the Vera Rubin platform have entered production. The Groq 3 LPX plays a role in complementing latency-sensitive intervals during the process of handling AI model inputs and generating responses. NVIDIA claims that this device can increase inference throughput by up to 35 times per megawatt for models with trillions of parameters. However, this figure may vary in actual data center environments. CEO Jensen Huang mentioned in his keynote speech at GTC 2026 that the Groq 3 LPX is expected to ship in the second half of the year, around the third quarter. In the Korean market, the connection with the supply chains of memory, servers, networks, and cooling components is drawing attention.
-- 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
Wells Fargo Raises Marvell's Price Target to $310, Assigns Overweight Rating
Alpha Chips Expands ASIC Orders in the U.S.
Semiconductor Silicon Wafer Prices Rise by Approximately 10%, Benefiting Taiwanese Companies like GlobalWafers
US 10-Year Treasury Yield Rises to 4.74%, Increasing Pressure on Risk Assets
NVIDIA Denies Plans to Launch AI Chips for China by Year-End
Tencent Chip Head Gao Jianlin Leaves to Start RISC-V AI CPU Venture
JPMorgan: SK Hynix May Add $130 Billion in Shareholder Returns
Marvell and Google Reach $12.2 Billion Warrant Collaboration
Tesla's Next-Generation Roadster May Debut This Month: Supercar with SpaceX Cold Air Thrusters Capable of 'Flying' Set to be Unveiled
Short Selling Balance in South Korea Reaches 19 Trillion Won, Up 14%
LG Electronics Plans to Launch Next-Generation Humanoid Robot in 2027
Microsoft Maia 200 Improves AI Inference Performance by 30%
Ming-Chi Kuo: TSMC's Rumored $1 Billion Apple Chip Stockpile Due to Memory Shortage is False
Apple Considering Adoption of ChangXin Memory DRAM
Goldman Sachs: Storage Costs of Nvidia's Vera Rubin Chip Account for Approximately 62% of BOM
Microsoft Plans to Release MAIA 300 AI Chip in September
South Korea Establishes 50 Trillion Won Semiconductor Fund and Trade Financing
Global Chip LOF Hits Limit Down, Guotou Silver LOF Plummets Over 8%
Apple Tests CXMT Memory Chips to Alleviate Shortage
Tesla's Terafab Project Launched in Texas, Aiming for Over 1 Terawatt of Computing Power
OpenAI Acquires Partial Patent Assets from Rain AI
U.S. Will Not Ban Chinese Open-Source Models for Now
AMD Begins Production of Helios, Claims 30% Advantage in Tokens per Dollar
TSMC Expands CoW Packaging Outsourcing, ASE and Other OSATs to Take on Capacity
Samsung Launches BV-NAND Technology with 58% Increased Storage Density
Huawei Semiconductor Chief Warns Nvidia's Chip Expansion Approaches Physical Limits
Microsoft Raises Xbox Prices to 749 USD Amid AI Chip Costs
Dark Side of the Moon Denies Hong Kong IPO Rumors, Bithumb Plans to Go Public in 2028
Foreign Capital Returns to South Korean Stock Market, Buying Samsung and SK Hynix
Morgan Stanley: Cloud Capital Expenditure Could Reach $1.3 Trillion by 2027
Wells Fargo Raises Marvell's Price Target to $310, Assigns Overweight Rating
Alpha Chips Expands ASIC Orders in the U.S.
Semiconductor Silicon Wafer Prices Rise by Approximately 10%, Benefiting Taiwanese Companies like GlobalWafers
US 10-Year Treasury Yield Rises to 4.74%, Increasing Pressure on Risk Assets
NVIDIA Denies Plans to Launch AI Chips for China by Year-End
Tencent Chip Head Gao Jianlin Leaves to Start RISC-V AI CPU Venture
