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
vLLM Recipes website has undergone a major redesign, offering optimized deployment solutions for models and hardware.
ME News Update, April 22 (UTC+8), the vLLM project announced that its community-maintained configuration library vLLM Recipes website has undergone a major redesign. The new website aims to help users quickly deploy and run large language models, with a key new feature providing clickable answers to questions like “how to run a specific model on specific hardware to accomplish a particular task.” The site adopts a URL style similar to HuggingFace, allowing users to jump directly to optimized solution pages by replacing the domain part of the model URL. The platform offers optimized \vllm serve\ command-line instructions for various models such as Qwen3.6-35B-A3B, Kimi-K2.6, and hardware like NVIDIA H100/H200/B200/B300, AMD MI300X/MI325X/MI355X, among others. Users can browse solutions based on providers like Arcee AI, Baidu, ByteDance, DeepSeek, Google, Meta, Microsoft, and more. The platform is fully compatible with vLLM and provides links to official documentation, GitHub repositories, and a comprehensive list of model hardware compatibility. (Source: InFoQ)