Dataflare SDKs

Official, professional-grade SDKs for the Dataflare API. This monorepo contains high-performance, idiomatic clients for Python, TypeScript, and Go — optimized for data pipelines and AI-native applications.

---

What is Dataflare?

Dataflare is a structured data platform providing curated, searchable datasets via both REST and high-performance gRPC APIs. The SDKs in this repo make it easy to integrate Dataflare into any application — from data pipelines to LLM-powered tools.

---

Unified Features

Feature	Python	TypeScript	Go
REST Client	✅	✅	✅
gRPC Client	✅	✅	✅
Typed Models	Pydantic	Zod	Structs
Auto Pagination	✅	✅	✅
Exponential Backoff	✅	✅	✅
Memory-safe File Download	✅	✅	✅
Custom Error Classes	✅	✅	✅

---

Installation

Python

pip install dataflare-sdk

# With gRPC support
pip install "dataflare-sdk[grpc]"

TypeScript / Node.js

npm install dataflare-sdk

Go

go get github.com/dataflarelab/df-sdk/go

---

Authentication

All SDKs read your API key from the DF_API_KEY environment variable by default:

export DF_API_KEY="dfk_your_key_here"

You can also pass it explicitly at client construction time. See the per-SDK docs below for details.

---

Quick Start

Python

from dataflare import DFClient, AuthenticationError

with DFClient() as client:
    for doc in client.datasets.stream("legal", search_term="contract", limit=100):
        print(doc.title, doc.summary)

TypeScript

import { DFClient } from "dataflare-sdk";

const client = new DFClient();

for await (const doc of client.datasets.stream("legal", { limit: 100 })) {
  console.log(doc.id, doc.title);
}

Go

client := dataflare.NewClient()
docs, nextCursor, err := client.Datasets.Query("legal", dataflare.QueryOptions{Limit: 100})

---

Per-SDK Documentation

• Python Documentation
• TypeScript Documentation
• Go Documentation

For full integration guides, architectural deep-dives, and interactive API references, visit the Dataflare Developer Portal.

---

REST vs gRPC — When to Choose What

	REST	gRPC
Best for	General use, simple integrations	High-throughput pipelines, low latency
Protocol	HTTP/1.1 + JSON	HTTP/2 + Protobuf (binary)
Streaming	Cursor-based pagination	Native server-side streaming
Setup	Zero extra deps	Requires [grpc] extra (Python) or @grpc/grpc-js (TS)

Note: The gRPC client uses Server Reflection — your Dataflare plan must support gRPC access.

---

License

MIT — Copyright © 2026 Dataflare Lab