Futures
Access hundreds of perpetual contracts
CFD
Gold
One platform for global traditional assets
Options
Hot
Trade European-style vanilla options
Unified Account
Maximize your capital efficiency
Demo Trading
Introduction to Futures Trading
Learn the basics of futures trading
Futures Events
Join events to earn rewards
Demo Trading
Use virtual funds to practice risk-free trading
CFD
Stock CFD Derivatives
US Stocks
Access real US stocks and ETFs
HK Stocks
Trade quality Hong Kong-listed stocks
Korean Stocks
SK Hynix
Real Korean stocks and top assets
Stock Futures
High leverage, 24/7 trading
Tokenized Stocks
Backed by real stock assets
IPO Access
Unlock full access to global stock IPOs
GUSD
3.8%
Mint GUSD for Treasury RWA yields
Stocks Activities
Trade Popular Stocks and Unlock Generous Airdrops
Launch
CandyDrop
Collect candies to earn airdrops
Launchpool
Quick staking, earn potential new tokens
HODLer Airdrop
Hold GT and get massive airdrops for free
Pre-IPOs
Unlock full access to global stock IPOs
Alpha Points
Trade on-chain assets and earn airdrops
Futures Points
Earn futures points and claim airdrop rewards
Promotions
AI
Gate AI
Your all-in-one conversational AI partner
Gate AI Bot
Use Gate AI directly in your social App
GateClaw
Gate Blue Lobster, ready to go
Gate for AI Agent
AI infrastructure, Gate MCP, Skills, and CLI
Gate Skills Hub
10K+ Skills
From office tasks to trading, the all-in-one skill hub makes AI even more useful.
Samsung Electronics Considers Outsourcing ‘Back-End Design’ of Google’s 2nm TPU I/O Die
Samsung Electronics is considering outsourcing the back-end design work for the input/output (I/O) die of Google’s 2nm-based tensor processing unit (TPU). The company is sounding out potential partners for assistance as growing foundry orders have stretched its internal workforce.
According to multiple industry sources on July 15, Samsung Electronics has recently been surveying its design solution partners (DSPs) regarding demand for a back-end design project involving the I/O die of Google’s 10th-generation, 2nm-based TPU, codenamed “Icefish.”
The TPU is Google’s proprietary AI accelerator. It is used to run Google’s AI models, including Gemini. Google is reportedly co-designing its 10th-generation TPU with MediaTek and plans to begin mass production as early as 2028.
The TPU consists of a compute processor and an I/O die. The compute processor is expected to be manufactured by TSMC using its 1.4nm process, while Samsung Electronics will manufacture the TPU’s I/O die using its 2nm process. The I/O die handles data transfers between the compute processor and high-bandwidth memory (HBM).
DSPs convert customers’ chip designs into manufacturable designs optimized for Samsung Electronics’ foundry fabrication facilities. This process is known as back-end design. It includes such steps as placing and routing logic circuits for physical implementation on a chip, designing test circuitry, and verifying the design. Samsung Electronics is considering whether to outsource some or all of this work to DSPs or handle it internally.
For Tesla’s 2nm autonomous-driving chip, Samsung Electronics handled the back-end design using its own personnel. Recently, however, the company has reportedly faced a shortage of available staff amid a surge in demand for its 2nm process.
Samsung Electronics is said to have recently secured Anthropic and DeepX as 2nm customers in addition to Google and Tesla. An industry source said, “Some of the orders that TSMC, Samsung’s competitor in the 2nm market, cannot accommodate due to capacity constraints are flowing to Samsung Electronics.”
ADTechnology and Gaonchips are currently being discussed as potential outsourcing partners for the back-end design. Both companies are preparing for or already carrying out 2nm projects using the same process technology as Google’s TPU.
However, the companies have not shown particularly strong enthusiasm. They are already engaged in major projects, and back-end design is essentially a service contract with relatively low added value. Instead, they prefer application-specific integrated circuit (ASIC) projects in which they oversee the entire process from design through tape-out. Nevertheless, they are reportedly leaning toward undertaking limited portions of the work to establish a track record in advanced 2nm projects for a major technology company.
Although the figures vary depending on the process technology and contractual terms, back-end design projects are generally worth tens of billions of won, while ASIC projects are typically worth hundreds of billions of won. For ASIC projects that guarantee mass-production revenue, the contract value can expand into the trillions of won.
ADTechnology is currently focused on “ADP620,” a 2nm central processing unit (CPU) project. Based on this project, the company aims to surpass KRW 1 trillion in annual revenue between 2028 and 2029.
Gaonchips is also preparing to participate in the Ministry of Trade, Industry and Energy’s “K-On-Device AI” project, which is valued at approximately KRW 800 billion. The company plans to develop a 5nm-based chip for advanced driver-assistance systems (ADAS) in collaboration with Hyundai Motor and other partners.
In addition to ADTechnology and Gaonchips, Alphachips is also being mentioned as a potential contractor. The company reportedly views the Google TPU project as a growth driver and is more eager to participate.