Modern, async-first HTTP client for Python · Zero external dependencies · Built on standard library

from reqivo import Reqivo

with Reqivo(base_url='https://api.github.com') as client:
    response = client.get('/users/octocat')
    print(response.json())

Or async:

import asyncio
from reqivo import AsyncReqivo

async def main():
    async with AsyncReqivo(base_url='https://api.github.com') as client:
        response = await client.get('/users/octocat')
        print(response.json())

asyncio.run(main())

Reqivo is a modern, async-first HTTP client library built entirely on Python's standard library. No external dependencies, pure performance.

Reqivo is not a requests replacement. It's a modern HTTP foundation.

Reqivo exists to bridge the gap between low-level socket/ssl usage and the need for a fast, modern HTTP client:

• Using only the standard library (Python 3.9+)
• Providing an async-first API with sync support
• Supporting modern protocols (HTTP/1.1, HTTP/2, WebSockets)
• Maintaining zero external dependencies for maximum portability

Reqivo is for:

• Developers who want async HTTP without heavyweight dependencies
• Teams building cloud-native applications
• Projects where dependencies matter (embedded systems, security-critical apps)
• Anyone seeking performance and simplicity

pip install reqivo

from reqivo import Reqivo

def log_request(method, url, headers):
    """Hook that logs every outgoing request and passes it through."""
    print(f"-> {method} {url}")
    return method, url, headers

# Fluent API: configure base URL, auth, and hooks in one chain
client = (
    Reqivo(base_url='https://api.example.com', timeout=10)
    .bearer_token('my-token')
    .on_request(log_request)
)

response = client.get('/data')
# -> GET https://api.example.com/data
print(response.status)  # 200
print(response.json())  # {"key": "value"}
client.close()

from reqivo import Session

with Session(base_url='https://api.example.com') as session:
    response = session.get('/data')
    session.put('/data/1', body=b'{"updated": true}')
    session.delete('/data/1')

import asyncio
from reqivo import AsyncReqivo

async def main():
    async with AsyncReqivo(base_url='https://api.example.com') as client:
        response = await client.get('/data')
        await client.post('/data', body=b'{"key": "value"}')
        return response.json()

asyncio.run(main())

from reqivo import Session

def file_chunks():
    with open('large_file.bin', 'rb') as f:
        while chunk := f.read(8192):
            yield chunk

with Session() as session:
    session.post('https://upload.example.com/files', body=file_chunks())

from reqivo import Reqivo

client = Reqivo()
ws = client.websocket(
    'wss://echo.websocket.org',
    auto_reconnect=True,
    max_reconnect_attempts=5,
)
ws.connect()
ws.send('Hello WebSocket!')
message = ws.receive()
ws.close()

• ✅ Zero external dependencies: Pure Python (Python 3.9+)
• ✅ Async-first design: Built for modern async/await workflows
• ✅ Sync support: Also works in synchronous code
• ✅ Unified Facade: Reqivo / AsyncReqivo as single entry points with fluent API
• ✅ Full HTTP methods: GET, POST, PUT, DELETE, PATCH, HEAD, OPTIONS
• ✅ Global config: base_url and default_timeout at session level
• ✅ Hooks system: Pre-request and post-response interceptors (sync and async)
• ✅ Streaming uploads: Chunked transfer encoding for iterables and file-like objects
• ✅ HTTP/1.1 support: Full protocol implementation with automatic redirects
• ✅ WebSocket support: Sync and async clients with auto-reconnect and configurable frame limits
• ✅ Connection pooling: LIFO connection reuse with semaphore-based limits
• ✅ Type hints: Fully typed following PEP 561
• ✅ Memory efficient: Optimized with __slots__ on all core classes
• ✅ Comprehensive testing: 97%+ test coverage

• 🔄 Mutual TLS / Certificate Pinning (Planned v0.4.0)
• 🔄 Structured Logging & Metrics (Planned v0.5.0)
• 🔄 HTTP/2 support (Planned v1.0.0)

Note: Benchmark suite coming soon in a future release.

Comprehensive examples and documentation are available in the Examples Directory:

• Quick Start - Installation, basic GET/POST, async introduction
• Async Patterns - Concurrent requests, async patterns
• Session Management - Cookies, authentication, persistent connections
• Error Handling - Exception handling, retry logic, circuit breaker
• Advanced Usage - Connection pooling, streaming, WebSockets

# Run all tests with tox
tox

# Run specific Python version
tox -e py312

# Run with coverage
tox -e py312
coverage html && open htmlcov/index.html

• Python 3.9 or higher
• No external dependencies for core functionality

We welcome contributions! See CONTRIBUTING.md for guidelines.

MIT © 2026 — Rodrigo Ezequiel Roldán
View full license

• CHANGELOG - Version history and release notes
• GitHub Issues - Bug reports and feature requests
• Milestones - Planned features and development timeline