Futures
Access hundreds of perpetual contracts
TradFi
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
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.
GateRouter
Smartly choose from 40+ AI models, with 0% extra fees
Just been looking at AMD's recent moves and there's actually some interesting stuff happening beneath the surface that a lot of people are glossing over.
So AMD's data center revenue hit 5.4 billion in Q4, up 39% year-over-year. That's solid, but the real story is what's coming next. The company's basically making a strategic shift from selling individual accelerators to delivering full rack-scale systems called Helios. These aren't just GPUs—they're integrated platforms combining processors, accelerators, networking, and software stacks. The MI450 chips powering these systems are ramping up in the second half of this year, with OpenAI already locked into a multi-year deal for up to 6 gigawatts of deployment.
What's interesting is how AMD is positioning itself beyond just competing on raw GPU specs. Eight out of ten top AI companies are already running Instinct GPUs in production, which is a pretty significant adoption rate. But the real competitive moat here might be the systems approach rather than chip-by-chip competition.
Now, the CPU side is something I think gets underrated in these discussions. Epyc processors are seeing record server sales, and it's not just about traditional workloads anymore. As AI infrastructure scales, you actually need serious CPU optimization happening in parallel—orchestration, memory management, data preprocessing, handling those non-GPU tasks that keep systems running efficiently. The fifth-gen Turin and fourth-gen Genoa chips are seeing strong demand, and management expects this momentum to carry into Q1 despite seasonal headwinds. Venice CPUs launching in H2 2026 already have solid demand signals.
The third piece here is the software execution, which honestly has been AMD's traditional weak spot. But they're actually making real moves with ROCm, their open-source ecosystem. Day-zero support for Instinct GPUs, integration with popular inference engines, a full enterprise AI stack—this matters because software lock-in is how you build defensible market position. Better software means stickier customers and higher margins over time.
Stock's down about 12% despite solid earnings, which usually means the market had unrealistic expectations baked in. But structurally, AMD looks well-positioned for the next cycle. The rack-scale systems strategy, CPU optimization for AI workloads, and improving software ecosystem create multiple vectors for growth. Worth keeping on the radar if you're thinking about long-term tech exposure.