Skip to main content
Synpath: one API for every prediction market

Synpath – One API for prediction markets

Build on prediction markets without building against each one.
One open-source API across Kalshi, Polymarket and Polymarket US: market data, order entry, live streams, and an execution engine that remembers what it sent. In-process from Python, over REST and WebSocket from anywhere.

License: MIT Python 3.10+ Discord

🆕 New: Every Kalshi order book, every tick.
Full-depth books and trades for every past Kalshi trade. Query any moment in under a minute. Get the data →


Supported Exchanges

Exchange id Market data Order entry Streams Order books
Kalshi Kalshi kalshi ✓ ✓ WebSocket one book per market, both sides read it
Polymarket Polymarket polymarket ✓ ✓ WebSocket one book per outcome token
Polymarket US Polymarket US polymarket_us ✓ ✓ retail and exchange APIs WebSocket and gRPC one book per market, both sides read it

Why Synpath

  • Liquidity is fragmented. Your time shouldn't be. The same market trades on Kalshi, Polymarket and Polymarket US, each with its own API, units and quirks. Synpath gives you one interface for all of them, and smart order routing buys from whichever book is cheapest. Spend your time on alpha, not plumbing.

  • Traders deserve advanced order types. Stops, trailing stops, icebergs, OCO, brackets, TWAP and pegs, on every venue, even where the exchange has none. Orders are journaled before they're sent, so a crash never places one twice.

  • A home after Dome and pmxt. Dome's API shut down in April 2026, and pmxt hasn't shipped since July 2026. Synpath is MIT-licensed, actively maintained, and follows ccxt conventions.

Installation

pip install synpath              # market data, order entry, streams, the engine and the server
pip install "synpath[grpc]"      # + Polymarket US exchange gRPC streams

Python 3.10 or newer.

Hosted API. Run synpath login, then synpath keys create. It serves tick-level Kalshi order book and trade history, and cross-venue market matching. Trading stays on your machine: Synpath never holds your keys or funds.

Quick Start

Markets and quotes

import synpath

kalshi = synpath.Kalshi()
market = kalshi.fetch_markets(limit=1)[0]

print(market.title)                                # Will Elon Musk visit Mars before Aug 1, 2099?
print(market.yes.quote.bid, market.yes.quote.ask)  # 0.1 0.12

Order books and search

book = kalshi.fetch_order_book(market.id, depth=5)
book.best_bid, book.best_ask                       # best first on both sides
kalshi.fetch_order_book(market.id, side="no")      # what NO costs

markets = kalshi.fetch_markets(query="trump", limit=10)
fee = kalshi.fetch_fee_schedule(markets[0].id)
fee.estimate(price=0.50, contracts=100)            # 1.75

Same code, every venue

for venue_id in synpath.exchanges:                 # ['kalshi', 'polymarket', 'polymarket_us']
    with synpath.exchange(venue_id) as venue:
        page = venue.fetch_markets(limit=5)

client = synpath.Client()                          # or one client, routed by the id
client.fetch_market("polymarket:2252244")          # every id starts with its venue

Order entry

from decimal import Decimal
from synpath import KalshiTrading, OrderRequest, Side, load_credentials, require

async with KalshiTrading(require("kalshi", load_credentials())) as kalshi:
    order = await kalshi.create_order(OrderRequest(
        market_id=market.id, side=Side.BUY,          # buy takes YES, sell takes NO
        amount=Decimal("10"), price=Decimal("0.42"), # always the YES price
    ))

Live streams

from synpath import PolymarketMarketStream, BookEvent

async with PolymarketMarketStream() as stream:
    await stream.watch_order_book(["polymarket:2252244"])   # both sides of the market
    async for event in stream:
        if isinstance(event, BookEvent):
            print(event.market_id, event.side, event.best_bid, event.best_ask)

The engine, and the orders the venues do not hold

from synpath.engine import Engine, EngineConfig, RiskConfig
from synpath import OrderType

async with Engine({"kalshi": kalshi}, EngineConfig(journal_path="trading.db"),
                  risk=RiskConfig(max_order_contracts=Decimal("100"))) as engine:
    await engine.submit(OrderRequest(
        market_id=market.id, side=Side.SELL, amount=Decimal("20"),
        type=OrderType.TRAILING_STOP, stop_price=Decimal("0.40"),
        params={"trail": "0.03"}, book="alpha",
    ))

REST server

python -m synpath.server                           # http://127.0.0.1:8000/docs
curl "localhost:8000/venues/kalshi/markets?limit=1"

Trading has its own app, with per-account keys and an append-only audit log:

python -m synpath.server bootstrap --control control.db     # the first key
python -m synpath.server schema --trading --out openapi.json
npx openapi-typescript openapi.json --default-non-nullable false -o src/synpath.d.ts

Documentation

  • CLI: hosted data API sign-in and keys versus local venue trading server setup
  • Unified API: methods, data structures, identifiers, paging, status, errors
  • Capabilities: what each venue can and cannot answer
  • HTTP server: routes, envelopes, typed client generation, trading routes with per-account keys, mounting behind your own auth
  • Trading: credentials, Decimal money, order entry on Kalshi, Polymarket and Polymarket US
  • Streaming: WebSocket and gRPC books, trades, orders and fills as self-healing event streams
  • Engine: a journal written before anything is sent, pre-trade risk, a ledger, reconciliation, a kill switch, the order types the venues do not hold, buckets that route one order across venues, paper trading
  • TypeScript client: generate a typed client from the OpenAPI document, place through cancel without Python

Development

git clone https://github.com/Synpath-ai/synpath
cd synpath
pip install -e ".[dev]"

pytest              # offline, against recorded venue payloads
pytest -m live      # market data against the real venues
pytest -m demo      # order entry on a venue's demo environment

The demo tests place and cancel real orders on a demo exchange with the credentials in your environment, and leave the account flat. python -m synpath.trading doctor says which venues are configured without printing a secret.

Prior Art

ccxt set the conventions this library follows. pmxt got to prediction markets first.

License

MIT

Metadata

Release files for synpath 0.1.0

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Source distribution (sdist)

Source distribution for synpath 0.1.0
File Size Uploaded
synpath-0.1.0.tar.gz 1.1 MB Details

Built distribution (wheel)

Table of built distributions (wheels) for synpath 0.1.0
File Interpreter ABI Platform
synpath-0.1.0-py3-none-any.whl Python 3 none any Details

Total release size: 1.4 MB

Release files / synpath-0.1.0.tar.gz

Download URL synpath-0.1.0.tar.gz
Size 1.1 MB
Tags Source
SHA-256 checksum
How to use checksums
8a846f86a5b3485538eab971613744b65061b5178e7007f6ef8559b0aa50ec58
BLAKE2b-256 checksum
How to use checksums
39b4717db70ebc7e774d1837be01f48d743ca130885a9def8dbf62408f0807f5
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
Yes
Uploaded via twine/7.0.0 CPython/3.13.14

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

PyPI Publish Attestation

PyPI verified that this artifact, at this checksum, originated from the publisher listed below.

Signed by GitHub Actions, verified by PyPI on Sep 28, 2026.

Transparency log

Release files / synpath-0.1.0-py3-none-any.whl

Download URL synpath-0.1.0-py3-none-any.whl
Size 348.1 kB
Tags Python 3
SHA-256 checksum
How to use checksums
305091f7ef70781195c3ce5a29fb930a26adfefa744ed9daf7d587f1d3112bce
BLAKE2b-256 checksum
How to use checksums
90d7103a93a23ef8126b8382641b7e20041012c8808200748b2831c5bcbee59e
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
Yes
Uploaded via twine/7.0.0 CPython/3.13.14

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

PyPI Publish Attestation

PyPI verified that this artifact, at this checksum, originated from the publisher listed below.

Signed by GitHub Actions, verified by PyPI on Sep 28, 2026.

Transparency log

Release history Release notifications | RSS feed

0.2.0

2 release files

This release

0.1.0 This release

2 release files

Anthropic, PBC Visionary sponsor Bloomberg Visionary sponsor Hudson River Trading Visionary sponsor Meta Visionary sponsor NVIDIA Visionary sponsor Microsoft Sustainability sponsor Depot Continuous Integration AWS Cloud computing and Security Sponsor Datadog Monitoring Fastly CDN Google Download Analytics Sentry Error logging StatusPage Status page