LingBunny Quant · Quantitative Strategy Platform

Write strategies as code,
run them live.

Write in Python, backtest on built-in historical data, and connect to QMT, XTP or CTP through one interface. Research, validation and order routing share the same environment — no more stitching market data to a broker API yourself.

Strategy language
Python 3.8 – 3.12
Asset classes
A-shares · ETFs · Convertibles · Futures
Deployment
Cloud-hosted / local runtime
3 languages
Strategy languages · Python / JavaScript / Pine Script
4 gateways
Trading gateways · QMT / XTP / CTP / Simulated matching
3 asset classes
Assets · Equities / ETFs & convertibles / Futures
1 engine
Matching engine · Shared by backtest and live

Strategy performance · Equity curve

Reports settle daily. Export to CSV.

ILLUSTRATIVE
Strategy equity Benchmark Daily · adjusted
01 Capabilities

Research, validation and execution, in one environment.

Not four open-source tools bolted together — one pipeline from market data to backtest to matching to order routing.

Strategy development

Write Python in the browser

Completion, go-to-definition, run and debug in the same editor. Export the file and run it locally whenever you want.

Backtest engine

Order-by-order matching, not close-price approximation

Replays minute or tick data with slippage and commissions included.

Trading gateways

One interface, many brokers

Swap gateways without touching strategy code — only the binding changes.

QMT XTP CTP IB Simulated
Risk controls

Checked before the order leaves

Per-symbol limits, daily loss caps and available cash are enforced pre-trade.

Market data

One adjustment basis, one calendar

Backtest and live read the same data source, so curves actually line up.

Languages & deployment

Three languages, cloud or local

Strategy files are never locked in. Export, migrate, reproduce locally at any time.

Python 3.8 – 3.12 JavaScript Pine Script Cloud-hosted Local runtime
02 Workspace

Strategy, backtest and orders on one screen.

Watchlist on the left, chart in the middle, orders on the right. Switching between backtest and live keeps the layout — only the data source changes.

03 Why replace the stitched-together setup

The glue between backtest and live is usually where the money leaks.

One codebase for the backtest, another for live orders — and the moment matching logic, slippage assumptions or event ordering diverge, the pretty curve stops reproducing. LingBunny runs both through the same engine.

01

Backtest and live share the matching code

Placement, cancellation, partial fills and limit-up rejections all go through the same execution engine. There is no second implementation.

02

Market data and gateways are hosted for you

No storing ticks, no reconnect loops, no calendar or corporate-action maintenance. The platform handles it; your strategy only sees signals.

03

The code is yours and runs locally

Export strategy files and run them on your own machine. Hosting is available, but your strategy is never bound to the cloud.

04 Trading gateways

Supported gateways and instruments

What you can actually trade depends on the permissions your broker or futures firm grants you. The table lists what the platform supports.

GatewayMarketInstrumentsConnectionStatus
QMT A-shares Shanghai/Shenzhen equities, ETFs, convertibles Broker client Supported
XTP A-shares Shanghai/Shenzhen equities, options Broker counter Supported
CTP Futures Commodity and financial futures Futures counter Supported
SIM All markets Same engine as live Built in Supported
QMT A-shares · equities / ETFs / convertibles Supported
XTP A-shares · equities / options Supported
CTP Futures · commodity / financial Supported
IB Overseas · equities / futures Planned
SIM Simulated matching · all markets Built in
+ More broker interfaces Evaluating

Gateway availability depends on whether your broker or futures firm has enabled the corresponding interface for your account. Confirm with your account manager before connecting.

05 Against building it yourself

You can assemble your own. The cost shows up later.

The expense isn't only development — it's the daily consistency maintenance and incident handling.

Concern Self-assembled LingBunny Quant
Market data Integrate vendors separately You reconcile adjustment factors and halts yourself Single data source One adjustment basis and calendar
Backtest vs. live Two codebases Matching details drift; curves stop reproducing Same matching implementation Switching modes only swaps data and gateway
Reconnection Write your own watchdog and alerts Overnight drops are found the next morning Hosted by the platform Crashed processes restart and log automatically
Risk controls Patched in as you go Limits and cash checks get missed Enforced before every order Per-symbol caps and daily loss limits are configurable
Operations Servers, processes, logs on you Cloud-hosted, or fully local
Time to first run Weeks First backtest the same day
06 Getting started

From sign-up to your first live order, in three steps.

STEP 01

Connect an account

Bind a broker or futures account in the console and pick the matching gateway. Simulated accounts need no binding — usable right after sign-up.

STEP 02

Write and backtest

Write the strategy in Python, set the universe and date range, then run. Review equity, drawdown, turnover and fills.

STEP 03

Go live

Once the backtest checks out, deploy the same file to a live gateway. Orders, positions and fills appear in the console in real time.

07 Execution path

From one line of code to one fill, through a single engine.

Backtest and live share the matching implementation. Switching modes does not rewrite the strategy or change its parameters — only where quotes and orders come from.

01 · Research Strategy code Python 3.8 – 3.12 02 · Validate Backtest engine Historical data · simulated fills 03 · Execute Live engine Same matching implementation 04 · Fill Trading gateway QMT · XTP · CTP Same strategy file · no rewrite · only the data source changes
Live path Code reuse Date range, universe and adjustment are configured in the console
08 FAQ

A few things you may want to confirm first

Do I need to know how to code?

The strategy itself is Python. The platform ships templates and wrappers for market data, order placement and positions, so a dual moving-average strategy is about 20 lines. You don't handle tick storage, connection management or the trading calendar.

Can the platform see or reuse my strategy?

Strategy files live under your account and can be exported at any time. We don't use customer strategies for proprietary trading or provide them to third parties. If the code must never leave your machine, use local runtime — the cloud only provides data and gateway access.

Which brokers and futures firms are supported?

It depends on whether they offer QMT, XTP or CTP and have enabled it for your account. Most major brokers offer QMT; futures firms generally offer CTP. Confirm your permissions with your account manager before connecting.

Is backtest data the same as live market data?

Yes. Historical data used in backtests and real-time data used in live trading come from the same source, with identical adjustment and calendar handling — so results don't drift due to inconsistent data.

How is pricing structured?

Simulated backtesting is free. Live hosting is billed by the number of running strategies and excludes trading fees charged by your broker or futures firm. See the subscription page in the console for current pricing.

What happens if my machine or network goes down while a strategy runs in the cloud?

Cloud strategies don't depend on your local machine or connection. The platform monitors each process and restarts crashed ones automatically, with logs. Positions are not liquidated automatically in extreme cases — you'll need to confirm in the console.

Run the backtest first.
Then decide about live.

Simulated backtesting needs no account binding. Sign up and start.

No credit card · Unlimited strategies in simulation