llm-router

Name: llm-router
Availability: InStock
Author: AIsa

Route chat, vision, and comparison requests across 70+ AI models via one API key.

23|6|Updated Jan 30, 2026

Search Tags:#CLI #OpenAI-compatible #vision-analysis #model-comparison #unified-API #LLM routing #Python-client

Ready to run

npx skills add https://github.com/AIsa-team/OpenClaw-Skills --skill llm-router-aisa-team

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: llm-router
Source: https://github.com/AIsa-team/OpenClaw-Skills/tree/main/llm-router
Command: npx skills add https://github.com/AIsa-team/OpenClaw-Skills --skill llm-router-aisa-team

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

This skill provides a unified gateway to 70+ AI models via a single API key for chat, vision, and model comparison, simplifying integration for developers.

Core Features & Use Cases

Unified multi-model routing: send chat requests to OpenAI, Claude, Gemini, Grok, and more using a single endpoint.
Model comparison and vision support: compare model outputs and analyze images with a consistent API.
Quick Start: Use the Python client or CLI to access chat, vision, and comparison features with a single API key.

Quick Start

Set your API key: export AISA_API_KEY="your-key"
Basic chat: python3 scripts/llm_router_client.py chat --model gpt-4.1 --message "Hello!"
Vision analysis: python3 scripts/llm_router_client.py vision --model gpt-4o --image "https://example.com/image.jpg" --prompt "Describe this"
Model comparison: python3 scripts/llm_router_client.py compare --models "gpt-4.1,claude-sonnet-4-5" --message "Explain AI"
List supported models: python3 scripts/llm_router_client.py models

Frequently Asked Questions about llm-router

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

FAQPage Schema

How do I call multiple LLM models like Claude and Gemini using a single OpenAI-compatible API?▼

You can call multiple LLM models like Claude and Gemini using a single OpenAI-compatible API by using a unified API key. This gateway routes chat, vision, and comparison requests to 70+ models across diverse families through one consistent REST endpoint.

How do I compare outputs from different AI models side by side?▼

To compare outputs from different AI models, you can use the model comparison feature via CLI or Python client. By passing multiple model names and a message, the system queries each selected model and returns their responses together for direct evaluation.

Can I analyze images with different vision models without writing separate integration code?▼

Yes, you can analyze images with different vision models without separate integration code. The unified API provides a dedicated vision endpoint that accepts an image URL and prompt, allowing you to use models like GPT-4o through a consistent client interface.

What is the best way to manage API keys for routing requests across 70+ AI models?▼

The best way to manage API keys for routing requests across 70+ AI models is to use a single environment variable. By setting one key, you gain access to the unified gateway for chat, streaming, vision, and comparison tasks across all supported model families.

Does the unified LLM routing API support streaming responses for chat endpoints?▼

Yes, the unified LLM routing API supports streaming responses for chat endpoints. Alongside standard chat requests and vision analysis, the Python client and REST API provide streaming support to handle continuous data output efficiently.

How do I list all supported models available through the unified API gateway?▼

To list all supported models available through the unified API gateway, you can execute a simple command via the CLI. Running the models command returns the complete roster of 70+ accessible AI models for chat, vision, and comparison tasks.