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
The Dark Side of the Moon rewrites the terminal intelligent agent and renames it Kimi-code, fully aligned with the Claude Code architecture.
According to Beating Monitoring, the open-source terminal AI coding agent kimi-cli under Moon’s Dark Side is quietly undergoing repository migration and architecture rewriting, and has officially been renamed to kimi-code. To address the bottlenecks of the original Python version in interactive response and execution efficiency, the R&D team has fully shifted to the technical roadmap of Claude Code, a terminal tool under Anthropic, completing a full architecture overhaul based on TypeScript and the Bun runtime to achieve millisecond-level cold start and a smooth terminal user interface (TUI).
This architectural adjustment means Kimi has completely abandoned the original Python terminal technology stack and has fully benchmarked against and adopted Claude Code’s mature solutions. The tool uses Commander.js to perform command parsing in accordance with standardized rules, and it replaces Rich and prompt-toolkit with React Ink to implement a brand-new responsive TUI interface. The refactor involved 166 TypeScript source files, with a code increment of more than 3.8万 lines. In the SWE-bench Verified benchmark, the TypeScript refactored version based on the kimi-k2.5 model successfully solved 317 out of 500 development tasks (solution rate: 63.4%). While maintaining the performance level of the original Python version, it significantly improved stability and enhanced the network layer’s resistance to interference.
In addition to aligning the underlying architecture, kimi-code focuses on polishing the human–computer collaboration experience. The new version not only supports dragging video assets such as screen recordings into the terminal for multimodal analysis, but also deeply recreates several benchmark-grade design elements from Claude Code, including a “Plan Mode” that supports cursor-interactive editing, Emacs commonly used shortcut keys, a safety design for quickly exiting by double-clicking Ctrl + C, and support for connecting automated workflows through custom lifecycle hooks (Hooks). For compatibility across the multi-model ecosystem, kimi-code has opened up custom integrations for third-party large-model APIs, enabling the tool to serve not only within the Kimi family, but also as a unified terminal programming gateway across models.