Build 6060 of MetaTrader 5 shipped in July 2026 with native support for the Model Context Protocol and a bundled AI Assistant. Inside the terminal it reads a symbol or a trader's open positions and explains support, resistance and trend. Inside MetaEditor it writes, debugs and explains MQL5 code. A trader signs in free with an MQL5.community account, or plugs in their own OpenAI, Anthropic or Gemini key. That single release note is a better starting point for this comparison than any general claim about AI reaching trading platforms, because it is the first time a platform this size shipped AI as a built-in feature rather than a plugin.
This list only includes platforms with a published, dated AI feature we could verify on the vendor's own site. That makes it short: four entries, not ten. DXtrade, Match-Trader, TradeLocker and NinjaTrader are strong platforms in other ways, and if we mention them here it is to say plainly that we found no published AI feature to compare, and to check with the vendor directly before assuming otherwise.
AI trading platforms compared
| Platform | AI feature | Announced | What it actually does |
|---|---|---|---|
| eTrader AI | eTrader AI, block-based AI Traders | Live now | A trader wires blocks together (check the news, ask a model to decide, check a daily loss limit, then buy) using their own OpenAI, Anthropic or Gemini key |
| MetaTrader 5 | MCP support and AI Assistant, build 6060 | July 2026 | Analyzes symbols and open positions in the terminal, generates and debugs MQL5 code in MetaEditor |
| cTrader | Official MCP servers, AI Agent Connect, cTrader CLI | 2026 | Lets tools such as Claude Code, ChatGPT Codex, Cursor or Gemini CLI place trades and run analysis through natural-language prompts |
| TradingView | AI Chart Copilot | Public beta, 2 April 2026 | A browser side-panel that gives a technical breakdown, pulls news and fundamentals, and scans watchlists for setups |
1. eTrader AI
eTrader AI is where a trader builds an AI Trader the same way they would sketch a flowchart: drop in a block for a new candle, a block that checks the news, a block that asks a model such as Claude, Grok or Gemini to decide, a block that checks a daily loss limit, then a buy block. No code. The trader supplies their own model key, sets the rules and limits themselves, and can switch the AI Trader off at any moment. It runs in Trade mode, placing real orders, or in Signals only mode, where it pushes a notification instead of trading.
What makes this different from the other three entries on this list is that it was built as an AI-first automation path from the start, next to eTrader Code for traders who prefer to write Tide by hand. It is not a chat assistant bolted onto an existing terminal. The honest trade-off: eTrader is the newest platform here, and eTrader AI is newer still, so it has not built up years of shared configurations the way MetaTrader's robot library has. A trader who wants a battle-tested library of pre-built experts should keep that in mind.
2. MetaTrader 5
MetaQuotes describes build 6060's AI Assistant as working two ways. In the terminal, it looks at the symbol or position a trader has open and explains the technical picture in plain language. In MetaEditor, it writes MQL5 code from a description, debugs code that already exists and explains what a block does. Both paths accept a free MQL5.community sign-in, which uses MetaQuotes' own model access, or a trader's own OpenAI, Anthropic or Gemini API key for their preferred model.
The native Model Context Protocol support is the more consequential half of the release, because MCP is the connector standard that lets outside AI tools talk to an application in a structured way rather than through a chat window alone. Paired with the largest existing base of MQL5 robots and indicators, this gives MetaTrader traders a way to have an assistant reason about a strategy they already run, rather than starting from nothing.
3. cTrader
Spotware took the MCP idea further outward. Its official MCP servers, cTrader AI Agent Connect and cTrader CLI are built so that AI tools that already exist elsewhere, Claude Code, ChatGPT Codex, Cursor or Gemini CLI among them, can place trades and run analysis on a cTrader account through a natural-language prompt. Instead of Spotware building its own chat assistant inside the terminal, it opened a door for the assistants traders already use to reach in.
That is a meaningfully different design choice from MetaTrader's approach, and it fits cTrader's existing identity as a platform built around cTrader Algo in C#, a language developers were already comfortable scripting against. A trader or a developer at a brokerage wiring internal tools to cTrader now has an official, supported path to do it with an AI agent instead of custom code.
4. TradingView
TradingView's AI Chart Copilot went into public beta on 2 April 2026 as a side-panel inside the browser charting tool. It gives a technical breakdown covering moving averages, RSI, MACD, support and resistance and an overall technical rating, pulls in related news and fundamentals for the symbol on screen, and scans a trader's watchlist for setups that match.
TradingView is not a platform a firm runs its trading book on, brokers integrate with it for charting, so Chart Copilot is best read as an analysis layer rather than an execution feature. A trader using it for research still places the order somewhere else, on whatever platform their broker actually runs.
What we could not confirm
DXtrade, Match-Trader, TradeLocker and NinjaTrader are all real, documented platforms with strengths worth a separate look. We searched for a published AI feature on each and did not find one at the time of writing. That is not the same as saying one does not exist. If AI tooling is the deciding factor for your firm, ask the vendor directly and get the answer in writing rather than assuming either way.
A model key is the common thread across every AI feature on this list that a firm actually controls. eTrader AI, MetaTrader 5's AI Assistant and cTrader's MCP tools all work with keys the trader or firm supplies, OpenAI, Anthropic or Gemini. None of the four platforms above locks a trader into one AI vendor. If you are evaluating eTrader AI against writing your own MQL5 or cTrader Algo code by hand, our algorithmic trading basics guide covers the groundwork either path needs, and eTrader's product page has the current block library.
"An AI feature is only worth adopting if you control the model key and the limits. We built eTrader AI so a trader sets the rules once and can switch it off in one tap, not so a black box trades for them."
— The SGHK Team
Key Takeaways
- MetaTrader 5 build 6060 (July 2026) added native Model Context Protocol support and a bundled AI Assistant for market analysis and MQL5 code generation.
- cTrader's official MCP servers let outside AI tools such as Claude Code or Cursor place trades and run analysis through natural-language prompts.
- TradingView's AI Chart Copilot, in public beta since April 2026, is a research layer for charting, not an execution platform.
- eTrader AI lets a trader build an AI Trader from blocks using their own model key, with a daily loss limit and an off switch built in.
- DXtrade, Match-Trader, TradeLocker and NinjaTrader have no published AI feature we could verify; ask the vendor directly.
Frequently Asked Questions
Which trading platforms actually have AI features in 2026?
Four have a published, dated AI feature we could verify: eTrader AI (block-based AI Traders, live now), MetaTrader 5 build 6060 (AI Assistant and MCP support, July 2026), cTrader (official MCP servers and AI Agent Connect, 2026) and TradingView (AI Chart Copilot, public beta since April 2026). We did not find a published AI feature for DXtrade, Match-Trader, TradeLocker or NinjaTrader at the time of writing.
Do I need my own AI model key to use these features?
For eTrader AI, yes, the trader supplies their own OpenAI, Anthropic or Gemini key. MetaTrader 5's AI Assistant works with a free MQL5.community sign-in or the trader's own key to one of the same three providers. cTrader's MCP tools connect to whichever AI tool the trader or firm already uses.
Is TradingView's AI Chart Copilot a trading platform?
No. TradingView is a charting platform and community that brokers integrate with; it is not something a firm runs its trading book on. Chart Copilot adds technical analysis and news to the chart, but a trader still places orders on a separate platform.
Can an AI Trader in eTrader AI place real orders?
Yes, in Trade mode. In Signals only mode it sends a push notification and a note in the bot journal instead of placing an order, so a trader can watch its calls before switching it live.
Should I switch platforms just for an AI feature?
Weigh it against everything else the platform needs to do for your firm: instruments, dealing desk, CRM integration and the traders you already have. Book a call and we will walk through where eTrader AI fits next to what you run today.
About SGHK
SGHK is a FinTech company that designs and builds its own software for the trading industry: the eTrader trading platform, Launch your Broker and Launch your Prop Firm. Every product is written, hosted and supported in-house and licensed to trading firms, with the CRMs branded to them, all hosted by us in the cloud, managed by each firm and built to scale across clustered servers as our clients grow. Everything is encrypted, and each firm is the only one with access to its data and its clients' data.