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
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
Competitive Landscape Analysis — How Much Market Share Can AMD and ASIC Take?
Competition is intensifying, but the moat is deeper than expected.
NVIDIA's stock price repeatedly hits new highs, and the most common question from investors is: Will AMD and other companies developing their own chips take market share away from NVIDIA?
Let's look at AMD first. The MI300X and MI400 are already close to or even surpassing the H100/B200 in pure hardware computing power (TFLOPS), but AI accelerator competition is not just about hardware. The CUDA ecosystem has accumulated over 5M developers and thousands of acceleration libraries, while AMD's ROCm still faces compatibility and documentation issues. Most enterprises still prefer NVIDIA when deploying large models. Recently, there were rumors that Microsoft internally tested the MI400 for a non-critical inference task, but that was just a small order worth a few hundred million dollars, whereas Microsoft's annual orders alone exceed $20 billion.
Now, looking at ASICs. Google's TPU, Amazon's Trainium/Inferentia, and Meta's MTIA are indeed more cost-effective in certain inference scenarios, but their problems are: closed systems, specialized use, and slow iteration. A large model company that wants to use multiple chips simultaneously needs to maintain multiple compilation stacks and operational systems, which is very costly. Most customers ultimately choose a strategy of "mainly using NVIDIA + a few self-developed chips for diversification," rather than complete replacement.
According to SemiAnalysis data, NVIDIA's market share in AI training chips will be 92% in Q1 2026, and in inference chips 78%, remaining almost unchanged from a year earlier. Competition has not significantly eroded its share.
Therefore, I believe that within May, there will be no disruptive changes in the competitive landscape, and market sentiment will not worsen because of this. NVIDIA's stock price is still driven by its own growth expectations. Forecasting the price range at the end of May: $225–$240.
#Polymarket每日熱點