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.
Grok 4.5 tops Tencent Hunyuan Agent endurance leaderboard, with over 60% of tasks completed without a model
The benchmark includes 46 tasks, covering areas such as software engineering, experiment reproduction, scientific computation, and multimodal analysis. Each task runs for up to 90 minutes and usually requires continuous operation across hundreds of steps.
LHTB doesn’t only look at final success or failure. Completing part of a task also earns points based on the actual progress. Hidden verifiers check files, test results, and program output, and an Agent’s self-claimed completion doesn’t count.
The paper’s first release tested 15 models. GPT-5.5 completed 7 tasks, ranking first. Each model performs one task, averaging about 9.9 million Tokens and taking about 85 minutes.
Current publicly available results have expanded to 20 models. Grok 4.5 topped the list with an average score of 0.505, completing 13 tasks. Out of the 46 tasks, 29 still have no model that meets the completion standard.
Most Agents can complete some steps, but they often run out of the 90-minute limit, or stop early before the task is finished, making it impossible to truly complete complex tasks.