openrouter-python-streaming

Stream OpenRouter LLM responses in Python using LiteLLM with async streaming and retry logic.

3|1|Updated Dec 23, 2025
One-click install
npx skills add https://github.com/marcus/marcus-skills --skill openrouter-python-streaming
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: openrouter-python-streaming
Source: https://github.com/marcus/marcus-skills/tree/main/skills/openrouter-python-streaming
Command: npx skills add https://github.com/marcus/marcus-skills --skill openrouter-python-streaming

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Connect to OpenRouter's unified API and stream LLM responses in Python, enabling real-time chat experiences and streaming insights without manual polling.

Core Features & Use Cases

  • Async streaming with LiteLLM and multi-provider access to 400+ models
  • Robust retry logic and backoff for transient failures
  • Production-ready usage patterns for building responsive AI assistants

Quick Start

Run the minimal example to start streaming responses from OpenRouter using LiteLLM.

Frequently Asked Questions about openrouter-python-streaming

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I stream LLM responses in Python using OpenRouter?▼

To stream LLM responses in Python using OpenRouter, this Skill implements async streaming via LiteLLM. It provides a unified API to access 400+ models and enables real-time chat experiences without manual polling.

What do I need to set up async streaming with LiteLLM and OpenRouter?▼

You need an OpenRouter API key and the litellm Python package to set up async streaming. This Skill uses these dependencies to establish production-ready streaming workflows and access multi-provider models.

Does LiteLLM handle transient failures during LLM streaming?▼

Yes, LiteLLM handles transient failures during LLM streaming through built-in retry logic and backoff. This Skill leverages these mechanisms to ensure robust streaming chat experiences when temporary network or API errors occur.

Can I access multiple LLM providers through a single Python API?▼

Yes, you can access multiple LLM providers through a single Python API using OpenRouter. This Skill supports multi-provider access to 400+ models via a unified API, allowing you to explore and integrate various models seamlessly.

What is the best way to build responsive AI assistants with real-time text generation?▼

The best way to build responsive AI assistants is by implementing async streaming with LiteLLM. This Skill provides production-ready usage patterns for real-time text generation, enabling immediate feedback and robust chat experiences.