DEV Community

linou518
linou518

Posted on

A-Share Quant Trading Platforms: QMT vs PTrade vs MyQuant — Which One Should You Choose?

Introduction: The First Roadblock in Quant Trading — Platform Choice

When most people decide to start quant trading on China's A-share market, their first search for "A-share quant trading platform" returns a flood of names: QMT, miniQMT, PTrade, MyQuant, xtquant, Hundsun…

These platforms appear to do the same thing, but their positioning is fundamentally different. Choose wrong, and you'll either waste time migrating later or, worse, put your strategy security at risk.

This article uses 2026 platform data to break down the three most popular options and answer: for each stage and need, which tool is the right fit?


The Three Platforms at a Glance

Before diving into details, remember these three lines:

  • QMT/miniQMT = Execution powerhouse. Sub-millisecond latency, local data, highest strategy security
  • PTrade (Hundsun) = Cloud-hosted lightweight option. Fast onboarding, no environment management needed
  • MyQuant (掘金量化) = Research-first platform. Best data quality for factor research and strategy incubation

These three aren't competing — they cover different phases of the quant development pipeline.


QMT/miniQMT: The Go-To for Technical Quant Traders

Key Strengths

Sub-millisecond execution, fully local

Single-trade latency under 1ms, full in-memory architecture, tick data delivered at microsecond speed. Critically: market data is cached locally, strategy code never leaves your machine, multi-layer encryption throughout.

For valuable strategies, local execution is the most important moat.

Widest instrument coverage

Equities, futures, options, margin trading, HK Stock Connect, convertible bonds, ETFs — all in one platform.

miniQMT mode: developer paradise

Via the xtquant Python package, you can call QMT interfaces directly from Jupyter Notebook or VSCode, bypassing the GUI entirely. For Python-native developers, the experience is far superior to any domestic competitor.

# miniQMT example: subscribe to real-time tick data
from xtquant import xtdata

def on_data(data):
    print(data)

xtdata.subscribe_quote('000001.SZ', period='tick', callback=on_data)
xtdata.run()
Enter fullscreen mode Exit fullscreen mode

Caveats

  • Windows 64-bit only — no Linux or Mac support
  • Requires account opening at a partner broker (e.g., Guojin Securities) before access
  • GUI is dated, but miniQMT mode rarely needs the GUI

PTrade (Hundsun): The Cloud-Hosted Lightweight Option

Key Strengths

Zero infrastructure management

Strategies run in the cloud, 24/7, without you maintaining a server. For users whose strategies are relatively simple and who want to minimize operational overhead, this is the biggest win.

Ready-made strategy templates

Stop-limit chasing, grid trading, VWAP order splitting — over a dozen components built in. Users without deep coding skills can go live by tweaking template parameters.

Caveats

  • Strategy code runs in the cloud → strategy leak risk for high-value proprietary strategies
  • Less Python customization depth than QMT
  • If you have Python skills, a self-hosted VPS + miniQMT setup is strictly more flexible

MyQuant (掘金量化): Best Data Source for the Research Phase

Key Strengths

10+ years of historical tick data

This is MyQuant's core differentiator. Nearly a decade of daily/minute/tick history across equities, futures, and fundamental/industry data — a full quality tier above most free data sources.

For factor research, high-quality backtesting data is the foundation of everything.

Multi-language support

Python, C++, and Matlab all supported — researcher-friendly by design.

Caveats

  • Live trading capability weaker than QMT (main advantage is the research phase)
  • Individual retail investor live trading channels are limited
  • Requires online connectivity; less local than QMT

Head-to-Head Comparison

Dimension QMT/miniQMT PTrade MyQuant
Focus Live execution Cloud-hosted live Research + live
Latency <1ms (lowest) Millisecond Not emphasized
Data security ⭐⭐⭐⭐⭐ ⭐⭐⭐ ⭐⭐⭐⭐
Historical data Moderate Moderate 10yr+, best quality
Instrument coverage Widest Equities + futures Equities + futures
Onboarding difficulty Medium Low Medium-High
OS support Windows only Cloud Cross-platform
Strategy security Highest (local) Lower (cloud) High

The Recommended Combination Path

The three platforms aren't mutually exclusive — the optimal approach is to combine them by phase:

[Research] MyQuant / qlib / backtrader
  → High-quality data, robust backtesting environment
    ↓
[Validation] MyQuant sim / QMT paper trading
  → Live-adjacent validation
    ↓
[Live Trading] miniQMT (technical) / PTrade (lightweight)
  → Deploy proven strategies
Enter fullscreen mode Exit fullscreen mode

Four-step path from zero to live quant trading:

  1. Register on MyQuant, run your first backtest (free, best data quality)
  2. Open a Guojin Securities account, apply for miniQMT access
  3. Connect via xtquant Python package, start with a daily-frequency strategy in paper trading
  4. Once stable, optimize for higher frequency and algorithmic order splitting

Conclusion: First, Figure Out Which Phase You're In

The core question in platform selection isn't "which is best" — it's "what phase am I in right now?"

  • Just starting strategy research → MyQuant: get your data and backtest infrastructure right first
  • Want to go live fast with a simple strategy → PTrade: zero-ops, quick start
  • Have Python skills, care about security and low latency → miniQMT is the long-term optimal choice

China's A-share quant regulations remain tight in 2026 (DMA restrictions, T+0 controls). Strategies must comply. But that doesn't stop you from building your toolchain today — so when a compliant window opens, you're ready to move.

Further reading: Quant factor research in practice (IC/IR/Barra multi-factor) / Python data sourcing for quant (tushare/akshare) / A-share risk control system design


Sources: CNBLOGS Quant Practice Series / MyQuant Official Documentation / 2026 Quant Community Synthesis

Top comments (0)