Does Keploy replace pytest + AsyncHTTPTestCase?

No. Keep pytest + AsyncHTTPTestCase for unit tests, where it is the right tool. Keploy covers the integration layer, where the cost is the dependencies rather than the assertions.

Do I have to change my Tornado code?

No. Keploy runs the process you already run and captures at the socket, so no annotation, no wrapper and no dependency injection change is needed.

What happens when a dependency changes?

Re-record. One command refreshes every captured interaction in the suite, so a changed contract shows up as a diff rather than as a silently passing stale mock.

Does it work with Tornado middleware?

Yes, and that is a reason to prefer recording here: middleware, authentication and serialisation all run on every recorded request, so the test covers the real request path rather than a handler called directly.

Can I run this in CI?

Yes, and that is the point: replay needs nothing running. No database service, no broker, no network access to upstreams, so the job is the same on every runner.

Keploy as a Tornado testing framework

Keploy tests Tornado services by recording real HTTP and database traffic from a running process, then replaying it as assertions. There is no test context to assemble and no framework wiring to change, because capture happens below the framework at the socket.

Generate Tornado tests free

keploy record -c "python app.py"

18.4K+

1.2M+300M+mocks createdSlack

What Keploy gives a Tornado team

Any Tornado service, unchanged

Keploy runs the Tornado process unchanged. Routing, middleware and serialisation all execute on every recorded request, because capture is at the socket rather than inside the framework.

Any Tornado version
Middleware runs unchanged
Routing and serialisation real
No plugin or adapter

Try Keploy for free Explore the docs

The problem

Why Tornado integration tests slow teams down

The friction is rarely the assertions. It is everything around them — spinning up dependencies, keeping mocks honest, and repairing tests after every refactor.

Book a call Explore the docs

Three ways to do it

Testing Tornado: by hand, with pytest + AsyncHTTPTestCase, or with Keploy

pytest + AsyncHTTPTestCase describes the exchange by hand and still needs the dependency supplied. Keploy records both halves once and replays them with nothing running.

By hand

pytest + AsyncHTTPTestCase

With Keploy

Select any row for the full comparison, with code.

Get started Book a call

Same coverage, three costs

What you write for Tornado vs what Keploy records

All three produce the same assertion. Only the third still passes after the next refactor without anyone editing it.

By hand2–4 hours

test_orders.py

hand-written

from unittest.mock import patch

from tornado_app import client

# Patching the repository means the Postgres driver never runs.

@patch("app.repository.find_order")

@patch("app.payments.authorize")

def test_confirms_an_order(authorize, find_order):

find_order.return_value = {"id": "o-1", "total": 4200, "status": "NEW"}

authorize.return_value = {"status": "AUTHORIZED"}

res = client.post("/orders/o-1/confirm")

assert res.status_code == 200

assert res.json()["status"] == "CONFIRMED"

The patches exist only for this test and are resolved by import path.

pytest + AsyncHTTPTestCase1–2 hours

test_orders_recorded.py

tool-assisted

import vcr

from tornado_app import client

# The cassette is cut on the first run, then frozen.

@vcr.use_cassette("cassettes/confirm_order.yaml")

def test_confirms_an_order():

res = client.post("/orders/o-1/confirm")

assert res.status_code == 200

# The Postgres half is not covered: a real instance is still required,

# seeded before the suite runs.

pytest + AsyncHTTPTestCase freezes the HTTP half into a committed file.

With Keploy~5 minutes

keploy/test-set-0/test-1.yaml

auto-generated

version: api.keploy.io/v1beta1

kind: Http

name: test-1

spec:

req:

method: POST

url: /orders/o-1/confirm

resp:

status_code: 200

body: '{"id":"o-1","status":"CONFIRMED","total":4200}'

assertions:

noise:

- body.confirmedAt

---

kind: Postgres

spec:

query: select id, total, status from orders where id = $1

rows:

- { id: "o-1", total: 4200, status: "NEW" }

---

kind: Http

spec:

request:

url: https://api.payments.example.com/v1/payment_intents/tok_123/confirm

response:

body: '{"status":"AUTHORIZED"}'

Recorded from one real request. Nothing here was typed by hand.

Times are estimates for authoring one endpoint’s coverage from scratch, not measurements.

Explore the docs Request a trial

Keploy vs the alternatives

Tornado testing tools, compared

The options a team on Python actually reaches for, and where each one genuinely wins. Select a row for the full comparison.

Keploy

pytest + unittest.mock

VCR.py

responses

Testcontainers

Postman

SmartBear ReadyAPI

Best in classStrongPartialNot covered

Assessments reflect each tool’s documented behaviour, not benchmark measurements.

Request a trial Book a call

How it works

Record your Tornado app once, replay it forever

Keploy sits below your Tornado process at the network layer. It watches the calls your app already makes, then serves them back on replay so tests run with no dependencies attached.

Your Tornado appGET/api/v1/orders/{id}200

RecordingKeploy proxyeBPF · userspace

PostgreSQLPostgres protocol beneath the ORM

MySQLMySQL protocol beneath the driver

MongoDBWire protocol beneath the Mongo driver

RedisRESP beneath the cache client

Outbound HTTPEvery outbound HTTP call

An example shape of a captured call. Your own endpoints and dependencies come from your real traffic, so nothing here has to be written by hand.

Try Keploy for free Request a trial

Mock coverage

What Keploy mocks for Tornado, with zero config

5 of the 5 dependencies a typical Tornado service talks to are stubbed from the recording itself — no mock classes, no fixture files, no containers in CI.

Auto-mocked
PostgreSQLView page
Postgres protocol beneath the ORM
Auto-mocked
MySQLView page
MySQL protocol beneath the driver
Auto-mocked
MongoDBView page
Wire protocol beneath the Mongo driver
Auto-mocked
RedisView page
RESP beneath the cache client
Outbound HTTP
Auto-mocked
Outbound HTTP
Every outbound HTTP call

Explore the docs Book a call

Quick start

Your first Tornado test suite in under five minutes

Every command below runs against your existing Tornado service. Nothing in your source tree changes.

1
1Install the Keploy CLI
A single binary. It needs a Linux kernel with eBPF support, or Docker on macOS and Windows — and it adds nothing to your project's dependencies.
curl -sSL https://keploy.io/install.sh | bash
2
2Record your service
Pass the command you already use to start the app. Keploy runs it and watches every socket it opens.
keploy record -c "python app.py"
3
3Exercise the paths you care about
Use curl, your frontend, or an existing smoke script. Every request becomes a test case with its Tornado calls captured alongside it.
curl -X POST localhost:8888/orders/o-1/confirm
curl localhost:8888/orders/o-1
4
4Replay in CI
Replay serves the recorded dependency responses, so the job needs no service containers and no Docker daemon.
keploy test -c "python app.py" --delay 15