"AlgoTest Zerodha" is a small search, but the person typing it is close to money. They already have a Zerodha account, they have seen AlgoTest, and they want to know one thing: what do I click, what will it cost, and what breaks? Most results answer a different question. They sell AlgoTest, or they explain Zerodha's Kite Connect API with no mention of AlgoTest at all.
This is the how-to I wanted when I first looked at this pairing. I'm Rajeev Gupta, founder of EliteAlgo. We run our own multi-leg NIFTY and SENSEX expiry-day option-selling strategies, so I read these setups as someone who has to live with the failures. I checked the prices and rules below against Zerodha's and AlgoTest's own pages on 8 November 2026. Where I could not confirm something, I say so instead of guessing.
Quick Answer
AlgoTest is a separate company. Zerodha does not own it and does not run it. You connect your Zerodha account to AlgoTest through Zerodha's Kite Connect API, and AlgoTest then places orders in your account when your strategy fires.
- What you need: an AlgoTest account, a Zerodha account with F&O enabled, and a Kite Connect app that you create yourself on Zerodha's developer site.
- What it costs on the Zerodha side: the Kite Connect Personal plan is free and covers order placement. The paid Connect plan is ₹500 a month per app and adds live and historical market data. Older articles that say ₹2,000 a month are out of date.
- What changed in 2026: since 1 April 2026, orders sent through the API must come from a static IP you have registered with Zerodha. Your home broadband will not do. AlgoTest sells a hosted machine with a fixed IP for this (they call it a Virtual Execution Environment).
- Also since 1 April 2026: AlgoTest no longer accepts market orders. Every entry and exit goes out as a limit order with a buffer you set.
- What it does not do: it does not make a bad strategy good. The quality of the backtest decides that, not the connection.
This page is analytical and educational. It is not investment advice, and nothing here tells you to trade any strategy.
Why People Search This Pairing
Zerodha is the broker most Indian retail traders already use. Kite is good for manual trading, but it does not backtest multi-leg option strategies and it does not run them for you. Streak, Zerodha's own strategy tool, covers some of that. If you want deeper options backtesting and a different execution style, you look at a third-party platform. AlgoTest is one of the most searched names for that. If you would rather stay inside Zerodha's ecosystem, our guide to Streak alternatives covers that route.
The appeal is simple. You keep your existing Zerodha account, margin and demat. You do not open a new broker relationship just to automate.
What You Need Before You Start
- A Zerodha account with F&O activated and the Kite developer login working. Kite Connect uses the same Zerodha client ID you trade with.
- An AlgoTest account, verified by email.
- AlgoTest credits. AlgoTest sells credits first and you spend them on plans. More on prices below.
- A static IP. Either your own (a fixed IP from your internet provider or a cloud server) or one that AlgoTest provides through its hosted environment.
- Your phone with the Zerodha TOTP app ready, because you will log in with it every trading day.
Kite Connect Pricing: Personal vs Connect
This is where most older articles are wrong, so here is the current list price from Zerodha.
| Plan | List price | What it includes |
|---|---|---|
| Kite Connect Personal | Free | Order management, GTT, alerts, margin calculation, portfolio management. No live market data and no historical data. |
| Kite Connect (paid) | ₹500 a month per app | Everything in Personal, plus real-time WebSocket data and historical candles. |
Two more details from Zerodha's own pages. The data APIs used to cost ₹2,000 a month and were cut to ₹500 from 6 May 2025. And from October 2026, if you are on the paid plan and generate at least ₹2,000 of brokerage in a month, Zerodha credits you 500 Developer Credits, which pays for the next month. Subscription fees are not refunded.
One thing I could not confirm: whether AlgoTest works fully on the free Personal plan or needs the paid data tier on your app. AlgoTest's own pricing article still says Kite Connect adds ₹2,000 a month, which is out of date. Ask on AlgoTest's Telegram support before you pay ₹500, and test with the free plan first. If your orders go through and your strategies run, you have your answer.
Step by Step: Connecting AlgoTest to Zerodha
AlgoTest has individual setup pages for most brokers. Its Zerodha page is no longer in the docs index when I checked, so the steps below combine the generic AlgoTest broker flow with Zerodha's own app-creation rules. Screens change, so treat this as a map and follow whatever AlgoTest shows you on the day.
- Open Broker Setup in AlgoTest. Click your profile icon at the top right, choose Broker Setup, and click the setup icon next to Kite Connect (Zerodha).
- Copy the redirect URL AlgoTest shows you. Do not type your own. This is the address Zerodha sends you back to after login, and it has to match exactly.
- Create your app on Zerodha's developer site. Log in at the Kite Connect developer console and create a new app. Give it any name. Paste the AlgoTest redirect URL into the redirect field and enter your Zerodha client ID. AlgoTest's earlier Zerodha guide said to leave the postback URL empty. If Zerodha offers a choice of plan at this step, start with the free Personal plan.
- Copy the API key and secret. They sit on your app's page under My Apps. Paste both into AlgoTest and save. Treat the secret like a password. Zerodha's docs say never to expose it in client-side code or share it.
- Register your static IP. In the developer console go to Profile, then IP Whitelist, and enter the IP your orders will come from. Details in the next section.
- Click Login in AlgoTest. You are sent to Zerodha. Enter your user ID, password and TOTP. On success you land back on AlgoTest with the broker marked as logged in.
- Test with one lot first. Run a single small strategy and watch the order book in Kite until you have seen an entry, a stop-loss and an exit go through cleanly.
If you see "The user is not enabled for the app", the forum threads I found all point to the same cause. You must log in with the same Zerodha client ID that you used when you created the app.
The Static IP Rule That Changed Everything in 2026
Under the retail algo framework from SEBI and NSE, Zerodha now accepts API order requests only from an IP address you have registered. This started on 1 April 2026. Zerodha's support page is clear on the details:
- It applies to order placement only. Market data streams, the order book and positions can still be read from any IP.
- You must register one primary IP. You may add a second as a backup.
- You can change the IP once per calendar week. Exchange rules set that limit.
- One developer account can cover family members. Zerodha lets you add the client IDs of your spouse, dependent children and dependent parents to separate apps. You confirm the IPs are used only by you and your immediate family.
Now the awkward part. AlgoTest runs your strategy and sends the orders. If the order leaves from AlgoTest's side, the IP that Zerodha sees is not your home IP. That is why AlgoTest built its Virtual Execution Environment (VEE). Their docs describe it as a dedicated Linux cloud machine with AlgoTest pre-installed and a fixed IP address, with 512 MB RAM and 10 GB storage. You buy it from the Broker Setup page for 30 or 90 days, and you can pay with credits, UPI, card, netbanking, wallet or pay later. They keep it reserved for three days after it expires. You can link brokers to it and buy more than one if you want separate setups, including for family accounts.
I could not find the VEE price on a public page, so I am not quoting one. Check the Buy VEE screen. Add it to your cost sheet, because on Zerodha it is now part of the running cost.
For wider background on how the rules work, read our SEBI algo trading regulations guide. In short, a personal algo that stays within 10 orders per second does not need its own exchange algo ID. Platforms that act as algo providers must register with the exchange, and they need their own static IP to connect to brokers. You do not register anything yourself when you use a platform, but you do own the account and its risk.
Your Daily Routine
This catches many first-time users. Zerodha's access token expires every day, at about 6 AM the next morning, because of a regulatory requirement. So even if the setup is a one-time job, the login is not. You open AlgoTest each trading day, click Login on the Zerodha broker and enter your TOTP. Do it before your first strategy is due, not at 9:14. Without a live broker session, AlgoTest has no way to place orders in your account.
Keep this in mind for expiry days, when your first entries may be scheduled within minutes of the open.
What AlgoTest Costs
AlgoTest runs on credits. At the base rate one credit is about ₹1. Bigger packs lower the per-credit cost, but AlgoTest's own pages show different pack prices in different places, so look at the Pricing button inside your account rather than trusting any article, including this one.
| Item | AlgoTest price (in credits) |
|---|---|
| Live execution, per strategy | 100 credits for 28 days (pro-rated if you add mid-cycle) |
| Forward test (paper trade), per strategy | 50 credits for 28 days |
| Free backtests | 25 per week |
| Extra backtests | 100 backtests for 100 credits, no expiry |
| Unlimited backtests, 7 days | 599 credits |
| Virtual Execution Environment | Price not published in the docs; see Buy VEE |
A rough monthly sheet for two live strategies on the free Kite plan: about 200 credits for execution, plus whatever the VEE costs, plus Zerodha brokerage. If you also need the paid Kite data tier, add ₹500. That is far less than the ₹2,000 plus platform fees older articles imply, but it is not zero, and slippage will cost you more than all of it.
Order Types: AlgoTest Is Limit-Only Now
This matters more for expiry-day sellers than the price list does. AlgoTest announced that market orders were disabled through the API. After 31 March 2026 the Market option was removed. A strategy activated with a market order after 1 April goes straight to an error state. Entries and exits must be limit orders with a buffer.
- AlgoTest recommends a buffer of at least 8 to 10 percent for options, and around 1 percent for cash and futures. A buffer that is too low can leave the order unfilled.
- If a limit order is not filled, the system chases the price for up to 50 seconds. After that it cancels the order, and the strategy goes into an error state. No open order is left in your book.
- Stop-loss exits and target exits are also converted to limit orders at your buffer.
Think about what that means on a sharp expiry-day move. A tight buffer can leave you stuck in a leg. A wide buffer fills but at a worse price, and the wider fill rarely shows up in a backtest. The Zerodha side agrees with this direction: Zerodha's API FAQ says market orders now need market protection and a value of 0 is rejected.
What AlgoTest + Zerodha Does Not Do
It does not validate your edge. A backtest is only as good as its fill assumptions. Ask how historical option prices are sourced and whether stops and entries assume the exact tick or a realistic fill. Our own walk-through on how to backtest an option-selling strategy shows the checks we run.
It does not remove margin risk. Short strangles and naked legs can use far more margin intraday than at entry. A clean equity curve can hide a day when the position nearly breached your available margin.
It does not guarantee live fills match the backtest. Between the AlgoTest server, the Zerodha API and the exchange, orders take real time. In the last hour of expiry, a few seconds is a lot. With the limit-only rule, the buffer you set becomes a cost you pay on both sides of the trade.
It is not an advisory service. It is analytical and execution software. You decide the strategy, and you own the result. Nothing here is a recommendation.
It does not replace your own stress test. Run any vendor backtest through at least one volatile stretch, such as a budget day or a global sell-off, before you trust it.
Common Errors and What They Usually Mean
| What you see | Likely cause |
|---|---|
| Orders rejected, API works for reading | The sending IP is not registered in Zerodha's IP Whitelist, or an IP change did not save (Zerodha allows one change per calendar week). |
| "The user is not enabled for the app" | You logged in with a different client ID than the one used to create the app. |
| Login works yesterday, not today | The Zerodha token expires daily. Log in again. |
| 403 or "insufficient permission" on market data | The app is on the free Personal plan, which has no data access. |
| HTTP 429 from Zerodha | You crossed 10 orders per second on one account. Orders that fail still count. |
| Strategy goes to error after a missed fill | The limit order was not filled within AlgoTest's 50-second chase. Widen the buffer, then re-test. |
Common Mistakes
Paying for the wrong plan. Many guides still tell you to buy a ₹2,000 data subscription first. Start on Personal, confirm it works for your use, and upgrade only if you need to.
Treating the static IP as optional. It is not. An unregistered IP means rejected orders on a day your stop-loss matters most.
Going live at full size. Run it one lot first, and run forward tests before that. A forward test costs 50 credits per strategy and uses no broker.
Sharing the secret. Your API secret is the key to order placement. Do not paste it in chats or screenshots.
Thinking automated means unattended. Check the order book yourself for the first several sessions, especially on expiry days.
Should You Build This Yourself Instead?
Some traders skip the platform and code against Kite Connect directly. With the free Personal plan, the API cost is zero, and you control every fill assumption. The trade-off is time, a server with a fixed IP, and the daily login, which you must script yourself. It is a legitimate route if you code, and our Kite Connect option-selling guide covers it. If you are choosing between brokers' APIs more widely, see our comparison of the Zerodha, Fyers and Angel One APIs.
Reading an AlgoTest Backtest Without Fooling Yourself
Suppose a NIFTY short-strangle backtest shows a high win rate and a positive monthly average. Check three things before you act on it. First, look at the worst single day and maximum drawdown next to the win rate. A short-premium strategy can win for months and then give most of it back in one session. Second, check that the test window includes volatile days, not just a long calm stretch. Third, look at return on margin, not return on notional capital. A figure calculated without margin looks much better than the same strategy measured against the SPAN and exposure margin you must actually keep. None of this is a complaint about AlgoTest. I apply the same checks to our own reports.
How This Compares With Other Options
This page is only about connecting AlgoTest to Zerodha. If you are still choosing a platform, the comparison belongs elsewhere. See Tradetron vs AlgoTest vs QuantMan for a side-by-side, and AlgoTest alternatives in India for other names. For the criteria we use ourselves, read how we evaluate platforms for option sellers. If you want to understand the strategy side, start from our NIFTY and SENSEX algo trading overview.
Frequently Asked Questions
Is AlgoTest owned by or partnered with Zerodha?
No. AlgoTest is a separate company. It connects to your Zerodha account through Zerodha's Kite Connect API, the same route other third-party tools use. I found no official partnership statement from either company, so confirm with both if that matters to you.
Do I need to pay for the Kite Connect API to use AlgoTest with Zerodha?
Zerodha's list prices are: Kite Connect Personal is free (orders, GTT, alerts, margin, portfolio, no market data), and the paid Connect plan is ₹500 a month per app with live and historical data. AlgoTest's own pricing article still says ₹2,000 a month, which is out of date. I could not confirm whether AlgoTest needs the paid data tier, so test on the free plan and ask AlgoTest support.
Do I need a static IP to use AlgoTest with Zerodha?
Yes, for order placement. Since 1 April 2026 Zerodha accepts API orders only from IPs you register under Profile, then IP Whitelist, in the Kite developer console. You can add one primary and one backup IP and change them once per calendar week. AlgoTest offers a hosted Virtual Execution Environment with a fixed IP for this.
Do I have to log in to AlgoTest every day?
Yes. Zerodha's access token expires at about 6 AM the next day, so you log in to the Zerodha broker inside AlgoTest each trading day, using your Zerodha password and TOTP, before your first strategy is due.
Does AlgoTest still support market orders?
No. From 1 April 2026 AlgoTest accepts only limit orders. It recommends a buffer of at least 8 to 10 percent for options. If a limit order is not filled within 50 seconds of chasing the price, it is cancelled and the strategy goes into an error state.
Is this pairing suitable for expiry-day (0DTE) option selling?
It can work, but expiry day is where limit-only orders, buffers and a few seconds of delay cost the most. Forward test first, then go live with one lot, and compare live fills with the backtest before adding size.
Is using AlgoTest with Zerodha investment advice?
No. AlgoTest is analytical and execution software. It does not tell you what to trade. You choose the strategy and you carry the risk. Nothing on this page is a recommendation.
Written by Rajeev Gupta, founder of EliteAlgo, an algo trading company operating in India since 2006 with a focus on intraday expiry-day option-selling research and execution. Prices and rules were checked against official Zerodha and AlgoTest pages on 8 November 2026 and can change, so confirm on the live pages before you pay or connect. This is an independent evaluation, not investment advice or an endorsement of any platform. See our SEBI algo trading regulations guide for the wider rules.
