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 30+ AI models, with 0% extra fees
Exa Upgrade Highlights: 500-character summary matches 8,000-character full text with high accuracy, tokens reduced by 16 times
CryptoWorld News reports that Exa AI Search API company has announced a major upgrade to the Highlights feature. Highlights is Exa’s ability to extract dense summary snippets related to a query from web pages. Each request generates them in real time (not cached), and completes within 100 milliseconds. In the SimpleQA benchmark test, 500-character Exa Highlights match original 8,000-character full texts with an accuracy rate, while token consumption is reduced by 16 times; the 4k-character Highlights perform better than 32k-character full texts.
The improvement is especially evident in long-document scenarios such as technical coding documentation, API references, SDK documentation, and research papers: with a 500-character budget, Highlights achieve 60% accuracy, while full texts are only 6%. Exa’s proprietary agent endpoints such as /answer, deep, and websets all use Highlights as the underlying retrieval mechanism and do not read the original page content.
This upgrade is at the model layer: all requests with the enableHighlights parameter enabled automatically benefit, with no code changes required.